strategy and supply chain management

STRATEGY AND SUPPLY CHAIN MANAGEMENT

RIRL 2010 - Bordeaux September 30th & October 1st, 2010

RIRL 2010 The 8th International Conference on Logistics and SCM Research BEM Bordeaux Management School September 29, 30 and October 1st 2010

Relations among supply chain design decisions and performance in push based contexts Margherita Pero Politecnico di Milano, Dept of Management, Economics and Industrial Engineering, Italy [email protected] Tommaso Rossi Carlo Cattaneo University – LIUC, Institute of Technology, Italy [email protected] Andrea Sianesi Politecnico di Milano, Dept of Management, Economics and Industrial Engineering, Italy [email protected]

Abstract This paper investigates the relations among supply chain design decisions (SCDD) and supply chain (SC) efficiency and effectiveness performance in a push based SC. SCDD are: the number of sources each node will buy from (multiple vs. single sourcing), the decision to install a certain level of production/inventory capacity at a certain node and to activate a certain number of nodes (splitting), the localization of the nodes (distance between nodes) and the number of levels of the SC. The SC performance analyzed are: average stock level of the SC and stock-outs at the retailer stage. A simulation model has been developed using ARENA software and results analyzed with statistical techniques (ANOVA). The relative magnitude of each SCDD has been figured out. These should be taken into account by SC managers when designing SCs in order to take proper action to reduce the negative effect. Key words: supply chain design, supply chain performance, push based contexts, simulation

1


1 : INTRODUCTION Researchers have investigated the determinants of supply chain performance, i.e. efficiency and effectiveness (Chase et al., 1963). These can be found both in the realm of supply chain management (SCM) and of supply chain design (Forrester, 1961; Sterman, 2000; Lee et al.,1997a; Lee et al., 1997b; Chen et al., 2000; Sezen, 2008). SCM is basically focusing on how to manage a supply chain once it has been built up, in order to reach target costs and service levels (Simchi-Levi et al. 2001). In this area, supply chain performance have been studied in relation to supply chain strategy definition (Fisher, 1997), business process connection (Krajewski et al., 2005), inventory planning (Ganeshan et al., 2001), client-supplier interaction (Salvador et al., 2001) and internal integration (Lee et al., 2007). Supply chain design studies focus on the decisions concerning the definition of the structure of a logistic network, so supply chain design does not include planning and control related decisions (Persson and Olnagher, 2002). Supply chain performance are strongly affected by supply chain design choices. Sezen (2008) demonstrates that “proximity to suppliers”, “number of suppliers”, “capacity planning along the chain” and “coordination of the logistic flow” significantly impacts the resources and output performance of the supply chain. The mathematical model developed by Helbing and Lammer (2005) uncovers the relation among supply chain structure and supply chain dynamics. In particular, since the relevance of topological features for the dynamic behavior of metabolic networks, food webs and cascade failures of power grids is commonly accepted, they assume that a similar relation is present in logistic networks too. In more detail, they develop a model focused on the dynamical property and linear stability of supply chains in dependence of the number of stages of the supply chain and the sourcing strategy of each node. Their model is based on conservation equations describing the storage and flow of inventories by a dynamic variant of Leontief’s classical input-output model and on a set of equations that reflects the delayed adaptation of the production rate to some inventorydependent desired production rate. They demonstrate that the logistic network topology, represented by the supply chain matrix (which synthesizes the number of stages of the supply chain, as well as the links among the supply chain nodes), has a direct influence on determining an over-damped or damped oscillatory behavior of the system expressed in terms 2


of the eigenvalues of the Jordan matrix corresponding to the considered supply chain matrix. Notwithstanding its relevance, the work of Helbing and Lammer (2005), on one hand, does not link the dynamical behavior of the network with the supply chain performance and, on the other hand, it considers only two supply chain design elements, i.e. number of stages of the network and sourcing strategies of each node. Simchi-Levi et al. (2001) state that also nodes’ capacity and localization are determinant factors of the performance of a supply chain, but they do not formalize nor validate their assumptions. Therefore, despite the proved importance of network topology in determining supply chain performance, little work has been done to investigate the relation among supply chain design decisions and supply chain performance (Bhatnagar and Sohal, 2005). Beamon (1998) outlines that few scholars study how the number of stages and the definition of which plant of the network will serve which customer, affect supply chain performance and none of them studies the two variables jointly. This paper aims at filling this gap in the literature. It investigates the relative effects of supply chain design decisions on supply chain performance by means of simulation techniques. The reason for the choice of simulation is almost twofold: (i) supply chain composition is an area where only early (even very interesting) proposals have been done and they are based, besides game theory (Cachon and Netessine, 2003) on simulation (Ding et al., 2005); (ii) with reference to network structure, several algorithms and linear programming models to support decisions in such area have been developed (Sridhar et al., 1999, Chopra and Meindl, 2001, Shapiro, 2001) but the majority of these models are intrinsically unable to consider both time dynamics and uncertainty. Procedures focused on simulation or on stochastic programming are more suitable for supporting decision making in an uncertain context (Alonso-Ayuso et al., 2002; Santoso, 2002). Here it is worth to notice that, due to the complexity of the subject, this work is focused on push-based logistic network. Pull-based supply chains have been studied in a previous work (Pero et al., 2010). In this paper, the following definition of push supply chain holds: each actor in a push-based supply chain bases its production decisions on forecast of the demand. Whereas, in pull-based supply chains, production is triggered by inventory level, i.e. when inventory reaches a certain level a certain quantity of product is produced. The paper is arranged as follows: section 2 introduces the background, while sections 3 and 4 present the methodology used and the obtained results respectively. In section 5 some concluding remarks and suggestions for future researches are reported. 3


2 : BACKGROUND 2.1: Supply chain design Supply chain design studies focus on one or more of the following three main areas (Shapiro, 2001): (i) supply chain composition, i.e. identification of entities (firms etc.) that should join the (constituent) logistic network; (ii) network structure, i.e. definition of the number of stages, of the sourcing strategies (multiple or single sourcing) of each entity and, finally, definition of the nodes’ capacity and localization (Simchi-Levi et al., 2001); (iii) collaboration level among the actors. With reference to the last area, it is worth to notice that some authors (Jagdev and Thoben, 2001; Makatsoris and Chang, 2004) consider collaboration level definition a supply chain design decision, whereas according to others (Simchi-Levi et al., 2001; Helbing and Lammer, 2005) the degree of collaboration required is a result of the decisions taken in term of network structure, i.e. configuration. In accordance with Simchi-Levi et al. (2001), we focus on configuration decisions.

2.2: Supply chain performance The topic of performance measurement has received increasing attention in the management accounting literature as well as in the SCM one (Cousins et al., 2008).

Supply chain

performance can be classified in efficiency and effectiveness measures (Beamon, 1999; Holmberg, 2000; Li et al., 2006; Tan et al., 1998). Efficiency refers to the ability of a firm to maximize the use of internal resources, given the same output. Efficiency measures are therefore related to costs, stock levels, machine saturation, productivity of internal resources. Effectiveness refer to the ability of a firm to satisfy clients requirements. Effectiveness can be measured, among others, against stock-out (or backlog) occurrences and quantities (Chase et al.,1973). Both kinds of performances are important. Indeed Beamon (1999) states that, when defining a supply chain performance measurement system, the main trade-off among costs and service level performance should not be neglected. In order to capture this fundamental trade-off, we consider both stock level and stock-out.

3 : METHODOLOGY 3.1: Variables and relationships The independent variables of our model are the decisions of supply chain design. As already stated, we will focus on the following decisions: 4


1. multiple sourcing: the number of sources each node will buy from. This means to decide whether to adopt a multiple or a single sourcing strategy; 2. splitting: this variable represents the decision of supply network design to install a certain quantity of inventory capacity at a certain node and to activate a certain number of nodes; 3. distance between nodes: this variable is connected to the localization of the nodes. We will express this variable in terms of distance between nodes; 4. number of levels (of the logistic network), e.g. a supply chain composed by n retailers and m manufacturers is a two-stage supply chain. The dependent variables of our model are supply chain performance, i.e. average stock level of the supply chain and number of stock-outs at the retailer stage. The aim of the study is to investigate the impact of supply chain design decisions on supply chain performance (see figure 1). Investigating the magnitude of such relations is beyond the scope of the work. As a consequence, an inductive approach can be used. Even if we reckon transportation and handling costs are relevant efficiency supply chain performance and quality, flexibility, responsiveness are relevant effectiveness supply chain performance to monitor, in line with the researches, among others, of Forrester (1961) and Sterman (2000), we focus on the supply chain performance stock-outs and stock level.

Multiple sourcing

Splitting

Supply Chain Performance • Average stock in chain • Stock-out quantities at the retailer stage

Distance between nodes

Number of levels Figure 1. The research model

3.2: Simulation model To investigate the relations, a simulation model has been developed by means of ARENA software. In the following the supply chain node behavior and the equations expressing it are presented. 5


We model supply chains composed of retailer(s), distributor(s) and manufacturer(s). The retailer, at the end of period t-1, (i) forecasts the expected demand for period t, (ii) compares inventory level with the expected demand, and (iii) places an order, Ot, to the upstream supply chain level. When the retailer receives an order, customer demand is satisfied demand with available stock. Every time the retailer inventory is not sufficient to satisfy the final customer demand, the retailer experiences a stock-out. Distributors and manufacturers, at the end of period t-1, (i) satisfy demand of downstream stage with available stock, (ii) if inventory is not sufficient, compute backlog, (iii) forecast expected demand for period t, (iv) compare inventory level with the expected demand, and (v) place an order, Ot, to upstream supply chain level. There is a minimum lead time between the time an order is placed and when it is received such that an order placed at the end of period t is received at t + d/s (if the supplier inventory is sufficient to satisfy the customer order), where d is the distance between supplier and customer and s is the average speed of the transport means. If the supplier inventory does not allow the customer order to be completely fulfilled a partial consignee occurs and the supplier backlog level increases. The manufacturer has limited production capacity whereas each actor has unlimited storage capacity and the manufacturer and the distributor have unlimited transport capacity. We represent the final customer demands at the retailer stage through the final customers inter-arrival time (drawn from an exponential distribution) and the quantity bought by the single customer (drawn from a normal distribution).

3.2: The model equations Each actor follows an order policy based on the expected demand, EDt , the available inventory and (when applicable) backlogs. The forecasting technique used by every node is exponential smoothing. In particular, in each period, t, for each actor, the expected demand, EDt, depends on the actual demand for the previous period, ADt-1, as well as on the expected demand for the past period, EDt-1 as in (1) EDt = α * ADt-1 + (1-α) * EDt-1

(1)

where α, whose value must range from 0 to 1, is the weight given to the actual demand in respect to the expected demand (in our model α = 0.5). 6


It should be noted that:

for the retailer: t is the sum of the length of the interval of time between two subsequent orders of the retailer to the distributor, i.e. in the model one week, and the transportation lead time from the distributor, i.e. in the model the rate between the distance and the speed of the transportation means;

for the distributor and the manufacturer: t is the length of the interval between two orders, i.e. in the model one week.

The order each actor places at the end of period t-1 for the period t, Ot, to the upstream supply chain stage is expressed by (2): Ot = MAX {0,O*t}

(2)

O*t = EDt - INVt-1

(3)

where, for the retailer (3) holds:

being INVt-1 the inventory level experienced by the actor at the end of period t-1 (i.e. at the beginning of period t). For the distributor and the manufacturer, who experience backlogs instead of stock-outs, the following equations (4) and (5) hold: O*t = EDt - INVt-1 + Bt-1

(4)

Bt-1 = MAX {0, ADt-1 - INVt-2}

(5)

where:

is the backlogs generated during the period t-1. It should be noted that in equations (3) and (4) inventory level is the physical inventory at disposal of the actor placing the order. The quantity already ordered but not available is not included in the formula. This is not needed since the set of parameters in the experimental campaign will be properly set. In particular, for each supply chain level the upstream delivery lead time is always (even in the worst case) negligible with respect to the fixed reorder frequency of the considered level, i.e. one week. For details please refer to the scenarios description in section 3.3. 7


For each period t, the retailer’s actual demand is calculated as the sum of the demands of the final customers arrived during the period t that he was able to satisfy (the retailer does not record stock-outs). For the distributor and the manufacturer the actual demand of each period t is given by the orders placed in the same period by the downstream stage. The quantity with which the distributor and the manufacturer provide the retailer and the distributor respectively in the period t is: Qt = min {INVt-1, ADt}

(6)

(such a quantity after d/s time units will increment the customer’s inventory). If at t the supplier’s inventory is not sufficient to completely fulfil the customer’s order, as soon as possible the supplier has to provide the customer with the pending quantity PQt,, computed as (7) PQt = ADt - INVt-1

(7)

As it can be noticed, the second term in equation (5) is the pending quantity.

3.3: The experimental campaign We used the ARENA™ software to simulate the above described supply chain model. We assume that: the duration of each period is 1 week; the mean of the exponential distribution followed by the final customers inter-arrival time is 1 hour; the mean and the standard deviation of the normal distribution followed by the quantity bought by the single customer are 4 and 1 units respectively; α is the same for each node and equal to 0.5; at each node the initial inventory is 2,000 and the expected demand of each actor for the first period is 0; the average speed of the transport means is 60 kilometres per hour; the simulation run length is 21 years and it is characterized by a 1-year warm-up period. Five types of supply chain topology scenarios have been defined: the first refers to the base case supply chain (i.e. one retailer, one distributor and one manufacturer), each one of the others refers to the base case supply chain where a single topological variable (i.e. number of stages, number of sources, distance between nodes and nodes capacity) is changed. For each scenario we modelled and simulated several configurations (totally 17), as depicted in table 1. Each configuration differ from the others for the value of the topological variable the scenario 8


refers to. Table 1 presents, for each topological scenario, the tested configurations, the final customer demand at the retailer and the manufacturer production rate. Here it is worth to notice that, concerning the topological variables ‘number of sources’ and ‘nodes capacity’, the analysis has been performed with reference to the retailer stage only as it is possible to see from table 1. Moreover, concerning the scenario connected to the ‘number of sources’, since we want that the installed capacity at each supply chain stage is coherent with the ones installed at the other stages, the number of retailers is always equal to the number of sources of each retailer, i.e. to the number of distributors. Again, in such a scenario, the different distributors have the same probability to be requested of products by each retailer.

9


Supply chain topology scenarios Name

Configurations Description

Base case: M

D

R

Distance: M

D

BC

Distance MD = 25 [km]; Distance DR = 400 [km]

D1

Distance MD = 212.5 [km]; Distance DR = 212.5 [km]

D2


D3


D4

Distance MD = 37.5 [km]; Distance DR = 600 [km]

R

R1

Di

Ri

M

MS1 MS2 MS3 MS4 MS7

Final customers interarrival time at R: ~ exp(1) Quantity bought by the single customer at R: ~ norm(4,1) Manufacturer production rate: 175 [units/h]

Two sources (i.e. i = 2) Three sources (i.e i = 3) Four sources (i.e. i = 4) Five sources (i.e. i=5) Eight sources (i.e. i =8)

Final customers interarrival time at each retailer: ~ exp(1) Quantity bought by the single customer at each retailer: ~ norm(4,1) Manufacturer production rate: 175 * i [units/h]

Four stages (i.e. i = 2) Five stages (i.e i = 3) Six stages (i.e. i = 4)

Final customers interarrival time at R: ~ exp(1) Quantity bought by the single customer at R: ~ norm(4,1) Manufacturer production rate: 175 [units/h]

a = 1/2 a = 1/3 a = 1/4 a = 1/5

Quantity bought by the single customer at each retailer: ~ norm (4,1) Inter-arrival time at each retailer ~ exp(1/a) Manufacturer production rate: 175 [units/h]

Number of sources: D1

Final customer demand and manufacturer production rate Final customers interarrival time at R: ~ exp(1) Quantity bought by the single customer at R: ~ norm(4,1) Manufacturer production rate: 175 [units/h]

Number of stages: M

D1

Di

R

Note: to comply with the fact that the supply chain length must be equal to 425 [km] the distance between the manufacturer and the first distributor is 25 [km], the distance between two subsequent distributors and between the last distributor and the retailer is 400/i [km] Nodes dimension (splitting):

NST1 NST2 NST3

R1 M

D R2

Note: the retailers ‘R1’ and ‘R2’ together must equal the retailer ‘R’ of the base case. As a consequence the following relations must be verified: R1 dimension = a * R dimension R2 dimension = b * R dimension a+b=1

ND1 ND2 ND3 ND4

Legend: R = retailer, D = distributor, M = manufacturer

10


Table 1 – Synthesis of the experimental campaign

For each simulation run we record, for each day, the quantity of stock-outs, i.e. number of units, and the quantity of backlogs, i.e. number of units, at the retailer and at the other actors respectively and the inventory levels, i.e. number of units, of each actor. In particular, the stock–outs quantity used for the analysis is the cumulative value over a simulated year (as a consequence for each simulation run 20 values of stock-out are available). The backlog quantity and the stock level used in the analysis are the average backlog per actor and the average stock level per actor over a simulated year.

4: RESULTS On the data gathered with the simulation, statistical analyses (basically ANOVA) have been performed for investigating the relationship between supply chain topology and supply chain performance as stated in section 3.1. Each decision variable’s effect on performance has been studied independently. It must be noted that the analysis of the splitting is still to be performed. It will be performed in the next future.

4.1: Distance When studying the effect on performance of the variable distance between nodes, two main analysis can be done: (i) the effect on performance of the positioning of the distributor with respect to the manufacturer and the retailer, and (ii) the effect on performance of longer supply chains. In particular, by comparing configuration BC, D1 and D2, the effect on performance of the positioning of the distributor with respect to the manufacturer and the retailer can be traced. Indeed, in the three scenarios, the whole distance between the factory, i.e. manufacturer, and retailer is 425 km, but the distance between factory and distributor varies. In particular, in configuration BC, the distributor is closer to the factory than in the others, vice versa in configuration D2 the distributor is closer to the retailer. In configuration D2 the distributor is in the middle between manufacturer and retailer. Indeed, in BD distance between Distributor and Retailer (Distance DR) is 400 km, and the distance between Manufacturer and Distributor (Distance MD) is 25 km. In scenario D1, Distance DR is 212.5 km and Distance MD is 212.5 km, while in scenario D2, Distance DR is 25 km and Distance MD is 400 km. The results of this analysis are discussed in section 4.1.1. 11


By comparing configurations BC, D3 and D4, the effect on performance of longer supply chains can be traced. Indeed, D3 and D4 are longer than BC, since in D3, Distance DR is 480 km and Distance MD is 30 km, while in D4 Distance DR is 600 km and Distance MD is 37.5 km. The results of this analysis are discussed in section 4.1.2.

4.1.1: The positioning of the distributor The ANOVA results show that the position of the distributor affects the stock-outs at the retailer stage and the average stock per actor (in both cases p-value=0 and residuals are normally distributed). In particular, when the distance between distributor and retailer decreases, i.e. moving from BC to D2, the stock-outs increase and the stock decreases. Table 2 shows the average values of the stock-outs in the different configurations. Configuration

Average Average stock stock outs per actor BC 1306 384 D1 1340 379 D2 1555 371 Table 2. Average values of the stock-outs at the retailer level and average stock per actor in the different configurations

The root cause of this results has to be sought in the way the topological variable ‘distance’ influences the expected demand at the retailer, i.e. the expected demand used to place the order to the distributor (see equations (1) and (3)). Each week the retailer computes the expected demand for period t. Period t is given by the sum of one week, i.e. interval between two orders fixed in all configurations, and the transportation lead time between retailer and distributor. In D2 the distributor is placed nearer to the retailer than in D1, but farther to the manufacturer. This has two effects: (i) on one hand, the requested quantity by the retailer in D2 is lower than in D1, (ii) on the other, the time the retailer has to wait if the distributor cannot satisfy retailer demand by means of its stock, is higher in D2 than in D1, since it is proportional to the distance between the distributor and the manufacturer. During this waiting time, the retailer can experience stock-outs. These problems might be solved by means of collaborative practices among retailer and distributor, e.g. sharing demand forecast and information regarding inventory levels.

12


4.1.2: The effect of the length of the supply chain The ANOVA results show that the length of the supply chain affects the average stock per actor (p-value=0.013 and residuals are normally distributed), whereas it does not affect the stock-outs (p-vale 0.882). However the stock-outs slightly increase. Table 3 shows the average stock-outs and stock per actor. Since the way according to which the retailer defines the quantity of the order to be placed upstream depends on the distance between retailer and distributor, the higher the distance the higher the order quantity, thus the stock. The slight increase in the stock-outs might be generated by the higher distance between distributor and manufacturer (see above for details of this effect). Probably the not statistical evidence for stock-outs is due to the fact that in all the three configurations the distributor is closer to the manufacturer than the retailer. Configuration

Average Average stock stock-outs per actor BC 1306 384 D3 1337 386 D4 1336 389 Table 3. Average values of the stock-outs at the retailer level and average stock per actor in the different configurations

4.2: Multiple sourcing When analyzing the decision variable multiple sourcing, the effect on supply chain performance of an increase in the number of sources can be traced. We hypothesized the relations among multiple sourcing and performance. In particular, we detailed the hypothesized relations among the independent variable ‘number of sources’, intermediate system variables and the dependent variable stock-outs at retailer stage, by means of a causal diagram. We outlined three main intermediate variables: (i) order variability at each single distributor, (ii) Forecast error of the distributor and (iii) stock at the distributor. On the relation among the number of sources and the order variability perceived by each distributor, the following can be stated. By construction (see table 1) when the number of distributors increases, the number of retailers increases accordingly. Let call n the number of retailers, i.e. of distributors. The probability that a retailer places an order to a generic 13


distributor is p=1/n. Each retailer places its order independently from what the other retailers do. Therefore, the probability distribution of the orders received by each distributor is a Binomial (n, p). When the number of distributors increases, i.e. of retailers, the variance of the demand experienced by each distributor, i.e. the orders placed on that distributor by the retailers, increases as well (Greene, 2000). Put differently, when the number of sources increases, the demand variability is expected to increase. Demand forecast accuracy at the distributor is negatively related to demand variability, i.e. the higher the variability the lower the forecast accuracy and therefore the higher the forecast error. When demand is underestimated, i.e. demand is higher than forecasted, retailers demand cannot be fulfilled with available stock, thus increasing the likelihood of stock-outs at the retailer. Vice versa, when demand is overestimated, i.e. demand is lower than forecasted, demand for the current period is satisfied and stock at the distributor increases thus reducing the probability of underfulfillment of retailers demand in the following periods (it plays the role of safety stocks). The hypothesized relations are depicted in figure 3. \

Number of sources +

Order variability at each single distributor

Forecast Error

Stock at the distributor -/+

+/-

+

Stock-out quantity at the retailer

Figure 3. Hypothesized relations among multiple sourcing and performance

The outputs of the experimental campaign have been analyzed in order to validate the depicted relations. Table 3 and 4 show respectively the average stock-outs (or backlog) quantities at each node of each level for each configuration, and the average stock per actor at each level for each configuration.

Average stockouts (or backlog) quantity At retailer level At distributor level At manufacturer level

Base Case

MS1

MS2

MS3

MS4

MS7

Direction of the effect

1306

984

918

897

873

847

↓

2249

8233

9389

9671

10099

10440

↑

3626

9129

11040

12859

14852

18791

↑

Legend: Base Case: Base case with one source, MS1: Multiple Sourcing with two sources, MS2: Multiple Sourcing with three sources, MS3: Multiple Sourcing with four sources; MS4: Multiple Sourcing with five sources; MS7: Multiple Sourcing with eight sources

14


Table 3. Average stock-outs quantity at each level depending on the configuration

Average stock quantity per actor

Base Case

MS1

MS2

MS3

MS4

MS7

Direction of the effect

At retailer level

231

227

226

225

225

224

Low ↓

At distributor level

460

546

569

576

587

595

↑

At manufacturer level

461

933

1403

1870

2340

3743

↑

Legend: Base Case: Base case with one source, MS1: Multiple Sourcing with two sources, MS2: Multiple Sourcing with three sources, MS3: Multiple Sourcing with four sources; MS4: Multiple Sourcing with five sources; MS7: Multiple Sourcing with eight sources

Table 4. Average stock quantity per actor at each level depending on the configuration

ANOVA analysis have been performed to check whether there is statistical evidence that the scenario affects the performance. Table 5 summarizes the results.

Effect Factor ANOVA p-value Stock-out at retailer Configuration 0 Backlog at distributor Configuration 0 Stock at distributor Configuration 0 Stock at retailer Configuration 0 Table 5. ANOVA analysis results (Configurations compared = BC, MS1, MS2, MS3, MS4, MS7)

Observed data are in line with what we expected: when the number of sources rises, stock at distributor increases (see table 4), while, stock-out at each retailer decreases (see table 3). Moreover, the increase in the stock at distributor accompanied by an increasing number of backlog at distributor is in line with the studies on Bullwhip Effect (Forrester, 1961). It is worth noticing that when the number of sources increases, the three actors in the supply chain experience different and contrasting effects: the retailers performance enhance, while distributors’ and manufacturers’ ones deteriorate. These results outline how, when the number of sources increase, supply inefficiency, e.g. higher stock, increases as well. Moreover, the fact that retailers experience better performance can become a hurdle to collaborate with the other actors in the supply chain.

15


4.3: Number of levels As it was expected, when the number of levels of the supply chain increases, given the same overall distance among the factory and the retailer, the quantity of stock in the chain increases, thus reducing the stock outs at the retailers. There is no reduction of the backlogs at the other levels since they are zero also in the Base Case (this is due to the parameters setting of the simulation model).

5: CONCLUSIONS Despite the proved importance of supply chain topology in determining supply chain performance, researchers neglected this relation. This work aims at filling this gap in the literature by investigating the relations existing among supply chain design decisions and supply chain efficiency and effectiveness performance in a push based supply chain. In particular, the analyzed supply chain design decisions are: the number of sources each node will buy from (multiple vs. single sourcing), the localization of the nodes (distance between nodes) and the number of levels of the supply chain. The supply chain performance analyzed are: average stock level of the supply chain and stock-out quantity at the retailer stage. To study these relations a simulation model has been developed using ARENA software. Five supply chain topology scenarios (the first scenario refers to the base case supply chain, each one of the others refers to the base case supply chain where one of the above mentioned topological variable is changed) have been defined along with 17 configurations derived from the scenarios. 17 simulation models representing the configurations have been built. Experimenting on such models allows to gather data on the performance of supply chain with different topologies. The analysis of the results of the experimental campaign by means of ANOVA has allowed us for outlining interesting remarks with reference to the topological variables ‘number of sources’. In particular, we noticed that when the number of sources increases, the three actors in the supply chain experience different and contrasting effects: the retailers performance enhance, while distributors’ and manufacturers’ ones deteriorate. It is interesting to notice the fact that retailers experience better performance can become a hurdle to collaborate with the other actors in the supply chain. We reckon that considering only stock level and stock-outs as performance variable is a limitation of our work. Therefore, future research will be aimed at investigating the impact of supply chain topological variables on transportation, handling and ordering costs.

16


REFERENCES o Alonso-Ayuso, A., Escudero, L.F., Garin, A., Ortuno, M.T., Perez, G. (2002), An approach for strategic supply chain planning under uncertainty based on stochastic 01 programming, Technical report, Departamento de Informática, Estadística y Telemática, Rey Juan Carlos University, Madrid, Spain. o Beamon, B.M. (1998), Supply chain design and analysis: Models and methods, International Journal of Production Economics, Vol. 55, n. 3, pp. 281–294. o Beamon, B.M. (1999), Measuring supply chain performances, International journal of Operations & Production Management, Vol. 19, No. 3, pp. 275-292. o Bhatnagar, R., Sohal, A.S. (2005), Supply chain competitiveness: measuring the impact of location factors, uncertainty and manufacturing practices, Technovation, Vol. 25, No. 5, pp. 443-456. o Chacon, G.P., Netessine, S. (2004), Game theory in supply chain analysis. In SimchiLevi, D., Wu, S.D., Shen, Z.J., (Eds.), Handbook of Quantitative Supply Chain Analysis: Modeling in the E-Business Era, , Kluwer Academic Publisher, Boston, Massachusetts. o Chase, R., Aquilano, N., Jacobs, R., (1973), Operations Management for Competitive Advantage, New York, New York: McGraw-Hill. o Chen, F., Drezner, Z., Ryan, J.K., Simchi-Levi, D. (2000), Quantifying the bullwhip effect in a simple supply chain: the impact of forecasting, lead times and information. Management Science, Vol. 46, No 3, pp. 436-443. o Chopra, S., Meindl, P. (2001), Supply chain management. Strategy, planning and operation, Prenctice Hall, Upper Saddle River, New Jersey

17


o Cigolini, R., Cozzi, M., Perona, M. (2004), A new framework for supply chain strategies: conceptual models and empirical test, International Journal of Operations and Production Management, Vol. 24, No. 1, pp. 7-41. o Cousins, P., Lawson, B., Squire, B. (2008), Performance measurement in strategic buyer-supplier relationships: The mediating role of socialization mechanisms, International Journal of Operations & Production Management, Vol. 28, No. 3, pp. 238-258. o Ding, H., Benyoucef, L., Xie, X. (2005), A simulation optimized methodology for supplier

selection

problem,

International

Journal

of

Computer

Integrated

Manufacturing, Vol. 18, No. 2-3, pp. 210-224. o Fisher, M. (1997), What is the right supply chain for your product?, Harvard Business Review, Vol. 75, No. 2, pp. 105-116. o Forrester, J.W. (1961), Industrial Dynamics, Pegasus Communications, Waltham, Massachussets. o Ganeshan, R., Boone, T., Stenger, A.J. (2001), The impact of inventory and flow planning parameters on supply chain performance: an exploratory study, International Journal of Production Economics, Vol. 71, No 1-3, pp. 111-118. o Greene, W. (2000), Econometric Analysis. Upper Saddle River, New Jersey: Prentice Hall. o Helbing, D., Lammer, S. (2005), Supply and production networks: from the bullwhip effect to business cycles, in: D. Armbruster, A. S. Mikhailov, K. Kaneko (eds.), Networks of Interacting Machines: Production Organization in Complex Industrial Systems and Biological Cells, pp. 33-66, World Scientific, Singapore o Hieber, R. (2002), Supply Chain Management – A collaborative performance measurement approach, VDF Hochshulverlang ag an der ETH 18


o Holmberg, S. (2000), A systems perspective on supply chain measurements”, International Journal of Physical Distribution & Logistics Management, Vol. 30 No. 10, pp. 847-68. o Jagdev, H. S., Thoben, K. D. (2001), Anatomy of enterprise collaborations, Production Planning and Control, Vol. 12, No. 5, pp. 437-451. o Lee, C.W., Kwon, I.G., Severance, D. (2007), Relationship between supply chain performance and degree of linkage among supplier, internal integration, and customer, Supply Chain Management: An International Journal, Vol. 12 , No. 6, pp. 444 – 452. o Lee, H., Padmanabhan, P., Whang, S. (1997a), The bullwhip effect in supply chains, Sloan Management Review, Vol. 38, No.3, pp. 93-102. o Lee, H., Padmanabhan, P., Whang, S. (1997b), Information distortion in a supply chain: the bullwhip effect, Management Science, Vol. 43, No 4, pp. 546-558. o Li, S., Ragu-Nathan, B., Ragu-Nathan, T.S., Rao, S.S., (2006) The impact of supply chain management practices on competitive advantage and organizational performance, OMEGA, Vol. 34 No. 2, pp. 107-124. o Makatsoris, H. C., Chang, Y. S. (2004), Design of a demand-driven collaborative supply-chain planning and fulfillment system for distributed enterprises, Production Planning and Control, Vol. 15, No. 3, pp. 256-269. o Pero, M., Rossi, T., Noé, C., Sianesi, A. (2010) An exploratory study of the relation between supply chain topological features and supply chain, International Journal of Production Economics, Vol. 123, pp. 266-278 o Perona, M., Miragliotta, G. (2004), Complexity management and supply chain performance assessment, International Journal of Production Economics, Vol. 90, No. 1, pp.103-115.

19


o Persson,

F., Olhager, J. (2002), Performance simulation of supply chain design,

International Journal of Production Economics, Vol. 77 , pp. 231-245. o Salvador, F., Forza, C., Rungtusanatham, M., Choi, T.Y. (2001), Supply chain interactions and time-related performances: An operations management perspective, International Journal of Operations & Production Management, Vol. 21, No. 4, pp. 461 – 475. o Santoso, T. (2002), A comprehensive model and efficient solution algorithm for the design of global supply chains under uncertainty, PhD thesis, School of Industrial and Systems Engineering, Georgia Institute of Technology, Georgia. o Sezen, B. (2008), Relative effects of design, integration and information sharing on supply chain performance, Supply Chain Management: An International Journal, Vol. 13, No. 3, pp. 233-240. o Shapiro, J. F. (2001), Modelling the supply chain, Duxbury, Pacific Grove, California. o Simchi-Levi, D., Kaminsky, P., Simchi-Levi, E. (2001), Designing and managing the supply chain, McGrow-Hill, Fairfield, Connecticut. o Sridhar, T., Ganeshan R., Magazine, M. (1999), Quantitative methods for supply chain management, Kluwer, Boston, Massachusetts. o Sterman J. (2000), Business dynamics: systems thinking and modeling for a complex world, McGraw Hill, Homewood, Illinois. o Tan, K.C., Kannan, V.R., Handfield, R.B. (1998), Supply chain management supplier performance and firm performance, International Journal of Purchasing and Materials Management, Vol. 34 No. 3, pp. 2-9.

20



Logistique inversée dans les retours de distribution entre grands distributeurs et fournisseurs au Brésil et en France Gisele de Lorena Diniz Chaves Université Fédérale du Espírito Santo (CEUNES/UFES) - Brésil [email protected] Rosane Lúcia Chicarelli Alcântara Université Fédérale de São Carlos (UFSCAR) - Brésil [email protected] Jacques Colin Université de la Méditerranée (Aix-Marseille II) - France [email protected]

Résumé Cette étude traite des flux inversés du détaillant vers leurs fournisseurs de produits carnés et laitiers au Brésil et en France de façon à répondre la question : comment les flux inversés de distribution sont-ils gérés par la grande distribution ? Cette recherche exploratoire a fait l’objet d’études de cas de dix sociétés françaises et brésiliennes. L’étude a mis en exergue une meilleure compréhension des pratiques de la logistique inversée et les résultats proposent certaines mesures de réduction des retours. Les produits périssables requièrent un grand contrôle sur les procédures, car plus influencés par les décisions commerciales comme les conditions de vente, les bonifications et les remboursements.

Mots clés: logistique inversée, retours de distribution, la grande distribution brésilienne, la grande distribution française, produits alimentaires périssables

1


1. Introduction La logistique inversée est une activité connue comme responsable de la planification et de la gestion du flux inversé de produits. Produits défectueux, périmés, recyclables, surplus d’une erreur de commande, entre autres, génèrent un flux contraire tout au long du canal de distribution vers la destination adéquate de chacun d’eux. Selon Rogers et Tibben-Lembke (2001), la logistique inversée coordonne les flux physiques et d’informations des retours des produits peu ou non utilisés (après vente), ou des produits usagés susceptibles d’être réutilisés, retournant vers le cycle de production ou de commercialisation. Dans le secteur alimentaire, Figueiredo et al. (2000, p. 104) ont constaté que les acteurs de la grande distribution brésilienne qui valorisent le mieux le service au client (comme critère de sélection des fournisseurs) sont ceux qui exigent le plus du service post-livraison. La gestion des flux inversés apparaît donc comme une alternative de différenciation des services offerts par la grande distribution. En outre, il est important de préciser que le résultat net de la grande distribution alimentaire en 2006 a été de 1,03 milliards de dollars, alors que le cumul des pertes a atteint près de 1,14 milliards de dollars selon l’Association Brésilienne de Supermarchés – ABRAS (Superhiper, 2008). Les pertes représentent moins de 1,97% du chiffre d’affaires des supermarchés, mais les supprimer doublerait les gains du secteur. Ces coûts sont principalement dus à des erreurs administratives, des fraudes ou à des marchandises invendables comme des produits périmés ou détériorés (ECR BRASIL, 2002). Contribuant en majeure partie aux ventes des distributeurs, les périssables (fruits, légumes, viandes, boulangerie, rôtisserie et frais) présentent le plus fort taux de pertes en raison de leur périssabilité élevée (Associação Brasileira de Supermercados, 2008). La courte durée de vie en rayon exige un taux de rotation élevé des produits et un contrôle rigoureux de la chaîne du froid, ce qui rend la gestion de cette catégorie de produits plus complexe. Dans le secteur alimentaire, les produits périssables nécessitent un système logistique efficient en rapport avec la faible valeur ajoutée de ceux-ci. De plus, le moindre problème dans la distribution des produits sous température dirigée, comme la variation de la température durant le transport ou le stockage, entraîne non seulement la modification des caractéristiques sensorielles du produit, mais aussi et principalement, un risque sur la sécurité alimentaire. En considérant que les nouveaux standards logistiques exigés par le marché créent le besoin d’une gestion efficiente et efficace des flux inversés pour les entreprises et qu’ils nécessitent des outils de planification et de contrôle de leurs opérations, la question qui oriente cette 2


recherche est : comment les flux inversés sont-ils gérés par la grande distribution ? Dans ce sens, cette étude cherche à comprendre non seulement la configuration du flux inversé des produits, mais aussi ses points critiques, les mesures de performance utilisées pour évaluer ce processus et, principalement, le rôle de la logistique inversée dans l’industrie alimentaire de façon à proposer une analyse de la gestion de cette activité par la grande distribution. Dans cette optique, suite à cette introduction, la deuxième partie propose un cadre conceptuel pour lànalyse de la logistique inversée, les aspects méthodologiques seront exposés dans une troisième partie, puis une quatrième partie présente les résultats et analyse les données recueillies. 2. La Logistique inversée 2.1 Définition et objectif de la logistique inversée La logistique s’est attachée à étudier tant les flux physiques que les flux d’informations et financiers, et ceci dans les deux sens de la chaîne d’approvisionnement : les flux directs et les flux inversés. La compétence logistique responsable de ce flux inversé dans la chaîne d’approvisionnement est appelée logistique inversée (Lambert et al, 1998). Elle est chargée de tous les retours des produits, des retours d’emballages, ainsi que de l’achat des matériels recyclables afin de les transformer de nouveau en matières premières. Ces dernières années, il a été constaté un intérêt croissant sur le sujet, tant par la littérature que par le marché lui-même et pourrait être défini comme le fait le Council of Supply Chain Management Professionals (2005): « un secteur spécialisé de la logistique orienté vers le déplacement et la gestion des produits et matériels après la vente et la post-livraison chez le consommateur. Ceci inclut les produits retournés pour réparation et/ou remboursement financier ». De manière plus aboutie et détaillée, Leite (2003, p. 16-17) décrit les attributions de la logistique inversée et ses avantages induit ainsi : [...] le secteur de la logistique d’entreprise qui planifie, opère et contrôle le flux et les informations logistiques correspondant au retour des biens de l’après vente et post-consommation vers le cycle commercial ou productif, par le moyen des canaux de distribution inversés, leur apportant une valeur de diverses natures : économique, législatif, logistique, d’image corporative, entre autres...

De Brito (2004) a proposé une approche de la logistique inversée dans laquelle les questions fondamentales de la logistique inversée sont posées : •

Pourquoi implanter ? – les motivations des entreprises quant à leur implication

dans la logistique inversée (stimulus) ;

3


•

Pourquoi retourner ? – les raisons selon lesquelles les produits sont retournés

(motifs du retour) ; •

Comment ? – comment le retour est-il réalisé ? (processus) ;

•

Quoi ? – qu’est-il retourné ? (caractéristiques du produit et types de produit) ;

•

Qui ? – qui réalise les retours ? (acteurs et rôles) ;

Les auteurs (Rogers; Tibben-Lembke, 1998; Fleischmann, 2001; Dekker et al., 2004) affirment que les objectifs pour implanter cette activité sont les suivants : économique, commercial, législatif et gains d’image. La plupart du temps, dans les entreprises, ces facteurs s’interagissent et ceci dans le but d’accroître leurs avantages stratégiques. Les facteurs motivant l’implantation d’une logistique inversée varient en fonction du type de produit ainsi que du type de retour : après-vente ou post-consommation. D’un autre côté, les retours postconsommation sont réalisés pour des motifs environnementaux et de récupération de la valeur du produit. Dans la phase d’après-vente, le retour se réalise pour des questions d’image du produit et de marque, satisfaisant par la même occasion les contrats d’approvisionnement ou la loi. La grande distribution et les grossistes valorisent ce service comme un élément de décision dans l’achat et l’évaluation de la performance de leurs fournisseurs en termes de disponibilité, de performance industrielle et de fiabilité. La logistique inversée augmente les coûts de changement de fournisseurs et fortifie la chaîne de valeur de l’entreprise. En tant que démarche, la logistique inversée assure la fonction de retour des flux de produits ou d’emballages. Néanmoins, il existe trois différents types de retours : les retours de fabrication, les retours de distribution et les retours du consommateur, comme le soulignent Dekker et al. (2004). Les retours de fabrication sont tous les retours identifiés durant la phase de production. Ils apparaissent pour plusieurs motifs, une matière première non conforme, une erreur dans le processus de production, un problème identifié par le test de qualité et jusqu’à un excès de matière première ou de produits. Les retours de distribution, comme leur nom l’indique, ont pour origine des rappels de produits, des ajustement de stock, des retours prévus par les contrats et des retours fonctionnels (Rogers; Tibben-Lembke, 1998). Ces derniers sont relatifs aux produits ou emballages qui ont pour particularité de faire l’aller et retour dans le canal, comme les palettes, et dont la fonction est de transporter d’autres produits en assurant cette finalité à plusieurs reprises. La logistique inversée est une vaste activité qui a pour commencement la collecte, l’inspection et le tri des produits et des emballages. Suite à la collecte, les produits subissent une analyse afin d’évaluer leur état général et leurs caractéristiques qualitatives. Suite à cette inspection, 4


les produits sont triés et distribués en accord avec le procédé de récupération déterminé. De cette façon, il est donc possible d’établir une série d’étapes à suivre : collecte J inspection J tri J récupération (Rogers; Tibben-Lembke, 1998; Fleischman, 2001 ; Dekker et al., 2004). 2.2 Facteurs critiques de la logistique inversée La prise en compte de la logistique inversée dans une stratégie d’entreprise fait partie d’un projet complexe car elle dépend de l’habilité dans la définition et l’implantation d’une forte intégration dans les systèmes et, parallèlement, elle implique une participation active des partenaires internes et externes. Dans ce sens, Lacerda (2003, p. 480) a identifié certains facteurs critiques pour une gestion performante de la logistique inversée, qui sont : •

De bons contrôles d’entrée;

•

Des processus cartographiés et formalisés ;

•

Des temps de cycle réduit ;

•

Des systèmes d’information performants ;

•

Un réseau logistique planifié ;

•

Une relation collaborative dans la chaîne ;

Avec de bons contrôles d’entrée, il est possible d’identifier correctement l’état des matériels retournés, afin que ceux-ci puissent suivre le flux inversé selon le procédé le plus adéquat ou, dans le cas contraire, d’interdire l’entrée des produits dans le flux. Des processus standardisés et cartographiés sont les conditions fondamentales pour exercer ce contrôle et pour obtenir les améliorations espérées, mais ils devront être révisés régulièrement. Afin de rendre performant le processus inversé, la réduction du temps de cycle des produits – le délai entre l’identification du retour jusqu’à la fin de son processus – (cycle time) doit être recherchée. Les temps de cycle trop longs peuvent être causés par de mauvais contrôles à la livraison, par un manque de structure physique et humaine spécialisée et par un manque de procédés clairs pour traiter les “exceptions”. Leite, Brito et Silva (2008) remarquent que, dans une étude réalisée sur 188 entreprises brésiliennes, le délai moyen pour compléter une opération de retour est d’une semaine au maximum. Les systèmes d’information permettent déjà d’obtenir des données essentielles pour le processus de retour des produits. Lee, Mcshane e Kozlowski (2002) affirment que de nombreuses entreprises utilisent le progiciel SAP pour la gestion des leurs opérations, mais la partie traitant de la logistique inversée s’avère insuffisante, puisque ce dernier à été conçu principalement pour gérer les flux directs de produits. Un système efficient doit intégrer des

5


variables relatives à la gestion des retours et posséder la flexibilité requise par làctivité, il doit donc être compréhensif dans ses spécificités et capable de fournir des données en temps réel. Pourtant, récupérer ou traiter ces systèmes d’information est une activité complexe au vu des variations et exceptions que génère ce processus. De plus, l’activité inversée requiert la définition d’un réseau logistique planifié, à savoir, une infrastructure logistique adéquate capable de traiter les retours du canal, incluant des centres de traitement, des moyens de stockage et des systèmes de transport. A plus grande échelle, il doit être envisagé une centralisation des réceptions et du traitement des retours. Le succès d’un système logistique inversé dépend de la capacité d’implantation d’une forte intégration entre les différents liens de la chaîne d’approvisionnement. Mais établir des relations collaboratives entre les agents n’est jamais simple. Il en ressort que cette relation est d’une extrême importance dans le secteur alimentaire où le moindre problème dans le canal compromet la qualité des produits et la santé des consommateurs. 2.3 Mesures de performance de la Logistique Inversée Une des raisons expliquant la nécessité de mesures spécifiques pour gérer le canal logistique est « la volonté de répartir les effets positifs et négatifs des évolutions fonctionnelles » à l’intérieur du canal de distribution (Lambert; Pohlen, 2002, p. 7). Selon Stank, Crum et Arango (1999), la relation entre la coordination inter entreprises et la performance de la chaîne d’approvisionnement doit être recherchée par la mesure du taux de service et du coût logistique, tels que : niveau de stock, coût de transport, coût du stockage, coûts de passation de commande, rupture de stock/disponibilité de produits, délais et régularité des échanges, ponctualité des livraisons, flexibilité, livraisons selon spécifications. Beamon (1999) propose trois types de mesures pour assurer le suivi de la performance de la chaîne d’approvisionnement : a) mesures des ressources incluant les niveaux de stock, le personnel, l’utilisation des équipements, l’énergie et les coûts ; b) mesures des résultats incluant le service au client, la qualité et la quantité du produit final fabriqué ; et c) mesures de flexibilité incluant d’adaptation au changement de volume et de délai des fournisseurs, fabricants et clients. Selon Lambert, Cooper et Pagh (1998), un bon audit doit comporter l’évaluation du marché externe (niveaux de service au client, conditions du marché et de la concurrence) et les opérations internes (service au client, transport, stockage, processus de commande, planification de la production et administration des stocks). La mesure de la performance réalisée par les entreprises de renommée internationale, selon le modèle du World Class Logistics (WCL), utilise des mesures de performance logistique selon 6


quatre dimensions : service au client, coûts, gestion des actifs et productivité (Global Research Team at Michigan State University, 1995). Ce modèle a pour particularité de proposer un format de lecture synthétique d’une logistique efficiente et efficace en relation avec les quatre domaines de compétence qui, articulés ensemble, rendent difficile toute reproduction par la concurrence. Hijjar; Gervásio; Figueiredo (2005) ont réalisé une révision bibliographique et décrivent les mesures de performance à l’intérieur de chaque dimension identifiée par le modèle WCL. Pour la gestion des flux inversés, il est intéressant d’analyser séparément les mesures utilisées pour évaluer cette activité. La révision de la littérature révèle peu de mesures de performance pour la logistique inversée, comme le montre le Tableau 1. COUTS Coût total de la logistique inversée Coût de prévention des retours Coûts d’erreur interne (coûts de mise au rebut, retouche, re-inspection, révision, obsolescence) Coût d’erreur externe (coût du processus de réclamation des clients, coût de retour des clients, coût de rappel des produits) Coût du processus de retour

Coûts des réparations et des procédés Coûts des litiges Rentabilité de la logistique inversée GESTION DES ACTIFS Obsolescence du stock

Fiabilité dans la gestion de l’inventaire SERVICE AU CLIENT Ponctualité des livraisons Casse Motif de retour Volume des retours

Réclamations (absence ou présence) Vitesse du retour Qualité d’expédition

Délai entre la réception du matériel au dock et sa disponibilité dans le stock (dock-to-stock time) AUTRES Conformité légale Niveau de satisfaction des autres participants de la chaîne d’approvisionnement avec des actions environnementales Restriction ou innovation dans la logistique

Lacerda et Ribeiro (2003); Sellitto et Mendes (2006); Chaves et al (2008) Rupnow (2006) Fassoula (2005) Fassoula (2005) Global Research Team at Michigan State University (1995); Bowersox et Closs (2001); Fassoula (2005), Hijjar et al (2005); Rupnow (2006); Matos (2007); Supply Chain Council (2008) Fassoula (2005), Rupnow (2006); Matos (2007) Chaves, Alcântara et Assumpção (2008) Richey et al (2005); Li et Olorunniwo (2008) Global Research Team at Michigan State University (1995); Abrahansson et Aronsson (1999); Bowersox et Closs(2001) Li et Olorunniwo (2008) Lacerda et Ribeiro (2003); Li et Olorunniwo (2008) Lacerda et Ribeiro (2003); Hijjar et al (2005) Hijjar, Gervásio et Figueiredo (2005); Chaves, Alcântara et Assumpção (2008); Supply Chain Council (2008) Bowersox et Closs (2001); Quintão (2003); Ramos (2004); Sellitto et Mendes (2006); Chaves, Alcântara et Assumpção (2008); Supply Chain Council (2008) Lacerda et Ribeiro (2003); Hijjar et al(2005) Lacerda et Ribeiro (2003); Li et Olorunniwo (2008) Global Research Team at Michigan State University (1995); Bowersox e Closs (2001); Lacerda et Ribeiro (2003); Li et Olorunniwo (2008) Li et Olorunniwo (2008) Sellitto et Mendes (2006) Sellitto et Mendes (2006) Richey et al (2005)

7

RIRL 2010 - Bordeaux September 30th & October 1st, 2010 inversée

Tableau 1 – Mesures de performance de la logistique inversée rencontrées dans la littérature Source : Elaboré par Chaves (2009, p. 89)

3. Méthodologie Les études traitant de la logistique inversée dans des secteurs spécifiques de l’économie sont encore récentes il est donc nécessaire d’approfondir les recherches de manière à mieux connaître ce sujet. La recherche de terrain réalisée pour ce travail est exploratoire, sans relation de cause entre les variables et reste descriptive, en abordant une étude de cas multiples, dont les données sont de nature qualitative. Les unités d’analyse de cette recherche sont constituées d’entreprises industrielles alimentaires de produits carnés et lactés, ainsi que d’acteurs de la grande distribution. Cette étude traite des flux inversés de produits de la grande distribution vers leurs fournisseurs de produits alimentaires sous température dirigée. Le choix des unités d’analyse s’est fait par échantillonnage non probabiliste de type intentionnel. Le critère de choix a été l’importance de l’entreprise ou sa contribution reconnue dans le développement de la logistique inversée, ainsi que sa disponibilité pour participer à la recherche. Les entreprises n’ont pas été identifiées pour des questions de confidentialité. Les données ont été collectées lors d’entrevues semi structurées par un questionnaire. La recherche sur le terrain a tout d’abord été réalisée en France en 2007 à l’aide d’études de cas dans trois entreprises du canal de distribution des aliments lactés : deux entreprises de production et un groupe de grande distribution. Malgré le nombre réduit d’entreprises qui ont participé à cette étude, chacune domine son secteur d’activité. Chaque entretien a duré en moyenne deux heures. Par la suite, la recherche sur le marché brésilien s’est réalisée dans le but de préconiser des recommandations pour la logistique inversée dans le canal de distribution agroalimentaire local. Entre juin 2008 et janvier 2009, sept entretiens individuels avec des entreprises ont été réalisés. Parmi ces sept entreprises, l’une produit uniquement des produits carnés, deux autres ne travaillent qu’avec le lait et ses dérivés, et la quatrième transforme les deux types de matières premières. Les trois autres entreprises appartiennent à la grande distribution. Les entretiens ont duré environ trois heures. Les acteurs de la grande distribution brésilienne ont été choisis parmi les plus importants, selon le critère de facturation du classement publié annuellement dans la revue spécialisée SuperHiper (2008). Les industriels alimentaires ont été choisis de la même façon à partir du classement de la revue « Revista Exame Melhores e Maiores » (Revista Exame, 2008). 8


Ces recherches n’ont pas pour but d’être comparatives. Le nombre de cas, le contexte économique, industriel et climatique, ainsi que l’objectif de la recherche dans les deux pays ne permettent pas de comparaison entre elles. L’objectif de la recherche en France a été de mieux comprendre la logistique inversée dans un pays avancé grâce à un stage de doctorat à l’étranger, mais aussi de servir de pilote dans l’étude du sujet au Brésil. Les études réalisées au Brésil ont eu pour but d’analyser l’opérationnalisation de la logistique avec l’intention de répondre à la question de recherche. La recherche en France a ainsi contribué à l’enrichissement de ce travail et à la découverte d’autres pratiques de gestion qui ont indéniablement interféré avec la recherche effectuée au Brésil. A partir de l’ensemble des données collectées, il a été réalisé une codification et une synthèse de celles-ci, qui permet d’identifier les plus pertinentes et de les confronter à la question de recherche et aux concepts développés dans la littérature. 4. Présentation des résultats et discussion 4.1

La Logistique Inversée en France vue par les entreprises étudiées

Dans les entreprises étudiées, la logistique inversée englobe le flux inversé des produits issus des retours des distributeurs vers leurs fournisseurs avant ou après le transfert de propriété. Les principales raisons de l’existence de ces retours sont dues à des erreurs d’expédition, aux produits endommagés durant le transport, aux produits défectueux ainsi qu’aux excès de stock dans le canal. Il est important de préciser que la fin de vie utile du produit n’est pas un facteur créateur de retours, la législation française et l’existence de la Fédération Française des Banques Alimentaires encouragent en effet la donation des produits s’approchant de leur fin de vie utile mais encore consommables. La relation entre les industries et la grande distribution est, en générale, collaborative sachant que les règles sont pré-établies, n’excluant toutefois pas l’apparition de conflits. La relation du groupe de la grande distribution avec ses fournisseurs est étroite : 70% des commandes sont établies en partenariat avec les fournisseurs (gestion partagée des stocks). De plus, le distributeur et ses fournisseurs partagent le même centre de distribution pour la livraison et la consolidation des chargements, leur permettant ainsi de diminuer leurs stocks et d’améliorer la ponctualité et la fréquence des livraisons, le tout ayant un impact sur la réduction des coûts. Le Tableau 2 synthétise les principales questions abordées. La logistique inversée dans le réseau de la grande distribution commence à partir de la réception des produits dans le centre de distribution. Afin de minimiser les retours, la centrale 9


d’achat du distributeur impose le retour systématique de la marchandise s’il est détecté une non-conformité avec la commande. A la réception le distributeur procède à tous les contrôles : si la livraison n’est pas conforme à la spécification de la commande (valeur, quantité, température, délai de péremption, avarie, respect de l’adresse et de l’horaire, aspect visuel, entre autres), les marchandises seront retournées au fournisseur. A partir du moment où le transfert de propriété de la marchandise est réalisé, ou autrement dit, dès lors que la réception accepte les produits, la responsabilité du produit appartient au distributeur. La seule exception sera dans le cas de problèmes de qualité du produit. A l’aide d’un système de gestion des stocks en magasin, le département de la qualité et de la sécurité alimentaire du distributeur est chargé de récupérer les produits à l’origine des pertes, mais aussi d’entrer en contact avec les organismes de charité. Etant donné que l’entreprise s’est engagée à retirer les produits du magasin une semaine avant sa date limite de consommation (DLC), les produits passent par une évaluation et s’ils sont considérés comme consommables, ils peuvent être acheminés vers la donation. Par ailleurs, ce distributeur a créé une dizaine de magasins (sans enseigne) où ces produits sont revendus à des personnes dans le besoin pour environ 10% de la valeur originale. Cette chaîne de magasins vend également des produits avec des défauts d’emballages. Questions

ENTREPRISE X

ENTREPRISE Y

ENTREPRISE Z •revente sur le marché d’origine; •revente sur d’autres marchés ; •donation; •réintégration au processus de fabrication; • mise au rebut • temps de cycle réduit • bons contrôles d’entrée • élaboration et révision des contrats entre industriels et distributeurs

Pratiques et processus

•revente sur le marché d’origine; •revente sur d’autres marchés ; •donation; •réintégration au processus de fabrication; •mise au rebut

•revente sur le marché d’origine; •revente sur d’autres marchés ; •donation; •réintégration au processus de fabrication; • mise au rebut

Facteurs Critiques

• temps de cycle réduit • relation collaborative dans la chaîne

• temps de cycle réduit • élaboration et révision des contrats entre industriels et distributeurs

• coût pour solutionner les litiges; • coûts de transport;

• coût pour solutionner les litiges; • coût des produits invendus ; • coût de la donation; • coût du rebut ;

• non informé

• élimination des produits impropres à la consommation ; • récupération de la valeur des produits

• élimination des produits impropres à la consommation; • récupération de la valeur des produits

• élimination des produits impropres à la consommation; • réduction des coûts de la logistique inversée

Mesures de Performance

Stratégie

Tableau 2 – Synthèse des principaux points de la recherche en France Sources : élaboré par les auteurs sur la base des résultats de la recherche

10


Pour l’entreprise Z, le bon contrôle à la réception des produits dans les centres de distribution, comme le montre la Figure 1, est le principal élément de la logistique inversée à être géré, puisqu’il s’agit du point où les retours peuvent être minimisés. En ce qui concerne les pertes et la casse dans les magasins, l’entreprise a établi un objectif (en pourcentage acceptable) qui est comparé mensuellement avec le résultat réel. Cet objectif est élaboré en fonction du budget, mais aucun chiffre n’a pu être divulgué. De cette façon, le processus maximisant la récupération de la valeur des produits est une priorité. Centre de distribution (gestion partagée des stocks)

Fournisseur

Analyse : le produit est-il conforme à la commande ?

NON

OUI

Restitution au fournisseur

Transfert de propriété

Problème de qualité

Produit proche de la DLC ou avec une avarie d’emballage ne compromettant pas la sécurité du produit

Produit avarié, expiration de la DLC ou autre cause rendant impossible sa consommation

Recall Vente dans des magasins populaires avec une réduction de prix

Donation

Mise au rebut par enfouissement ou incinération

Figure 1 – Représentation des flux de la logistique inversée de l’Entreprise Z Source : élaboré par les auteurs 4.2 La logistique inversée au Brésil dans l’optique de la recherche L’étude de la logistique inversée dans les entreprises sélectionnées a montré que chaque entreprise possède une motivation différente, comme lìndique au Tableau 3. Deux grands réseaux de distributeurs ne considèrent pas l’activité comme stratégique, même s’ils en reconnaissent l’importance de la gestion des flux inversés pour l’entreprise, mais sans pour autant y reconnaître un avantage compétitif. Le distributeur G considère que l’activité est stratégique pour la gestion de ses stocks, le fait est qu’elle a gagné de l’importance dans l’entreprise : il dispose de procédés planifiés et formalisés pour la logistique inversée, ainsi que d’un système d’information incorporant plusieurs variables inhérentes à cette activité. Du côté des industriels, la logistique inversée se révèle être stratégique pour chaque entreprise étudiée.

11

RIRL 2010 - Bordeaux September 30th & October 1st, 2010 INDUSTRI E A B

C D

ROLE STRATEGIQUE

ROLE STRATEGIQUE

DISTRIBUTION

maintien de la compétitivité garantit la sécurité alimentaire protège l’image de marque satisfaction du client garantit la sécurité alimentaire protège l’image de marque réduction des coûts réduction des coûts

E

non considérée comme stratégique

F

non considérée comme stratégique

G

permet de mieux gérer le contrôle des stocks

Tableau 3 – Rôle stratégique de la logistique inversée dans les entreprises étudiées au Brésil Source : élaboré par les auteurs Les caractéristiques spécifiques des aliments périssables définissent les processus les plus utilisés par la logistique inversée dans ce canal de distribution. Parmi les nombreux types de produits retournés, quatre catégories apparaissent en croisant les deux principaux processus (revente et rebut) avec type de transfert de propriété des produits (voir Tableau 4). Redistribution ou Revente Retour antérieur au transfert de propriété (Refus)

Produits sans avarie à l’origine d’une erreur d’expédition Produits sans avarie non-conformes à la demande

Retour postérieur au transfert de propriété (Restitution)

Produits sans avarie à l’origine d’un excès de stock dans le canal Produits saisonniers

Mise au rebut Produits possédant des problèmes de qualité visuelle et organoleptique Produits ayant subi des écarts significatifs de température Produits avariés pendant le transport Produits en fin de vie utile Produits possédant un problème de qualité intrinsèque Produits invendus

Tableau 4 – Catégories de retours des entreprises étudiées au Brésil Source : élaboré par les auteurs L’analyse des flux inversés des entreprises étudiées permet de mettre en évidence que lorsqu’un problème à la livraison a été identifié, c'est-à-dire avant l’acceptation des produits par le distributeur et le transfert de propriété, les accords commerciaux incluent une réduction du prix des produits ou une compensation des pertes dans le but de ne pas modifier le flux des produits incriminés par le distributeur, comme le montre la Figure 2. Identification du problème et restitution Retrait des produits du magasin Collecte par le fabriquant

Le distributeur se charge du rebut (en magasin ou CD)

Consolidation de la charge et stockage en CD Inspection par le fabriquant en CD

Collecte par le fabriquant Revente

Réparation Re-industrialisation

Rebut des produits au CD Rebut

Figure 2 – Flux inversés des restitutions des entreprises étudiées au Brésil 12


Source : élaboré par les auteurs Ces accords permettent que la marchandise soit acceptée même si elle présente certains problèmes (comme une date limite de consommation inférieure à la spécification de la commande) en échange de certains bénéfices, comme des réductions significatives, la garantie de dédommagement des pertes éventuelles, entre autres. D’un autre côté, lorsque le transfert de propriété est réalisé entre l’industriel et le distributeur, les flux inversés sont structurés de façon distincte. Le schéma des flux met ainsi en évidence la nécessité d’un nouveau transfert de propriété entre les entreprises étudiées dans le cas de retours des produits. La Figure 3 représente les flux de la logistique inversée entre le centre de distribution et les magasins des distributeurs.

Centre de distribution

analyse de la livraison

distribution vers les magasins

stockage

analyse des restitutions

rebut

Magasins

Rebut

Rebut

Rebut

retour sans transfert de propriété restitution après transfert de propriété rebut des produits par incinération ou enfouissement

Figure 3 – Flux inversés chez les distributeurs Source : élaboré par les auteurs Ainsi, à partir de l’étude des flux inversés des entreprises étudiées, il est possible de déduire leur structure illustrée dans la Figure 4. Les flux comprennent les retours sans transfert de propriété des produits ou refusés et les retours avec transfert de propriété des produits ou restitués.

Usines

CDf

CDf - centre de distribution du fabricant 13 CDd – centre de distribution du distributeur

CDd

Flux sans transfert de propriété Flux avec transfert de propriété

Magasins


Figure 4 – Flux inversés dans les entreprises étudiées au Brésil Source : élaboré par les auteurs Les motifs des retours influencent également les flux inversés, ils peuvent être organisés en fonction de leur origine : logistique, commerciale ou qualité. Le Tableau 5 révèle les raisons des retours mentionnés par les entreprises étudiées en fonction de leurs origines et causes.

Logistique avarie pendant le transport, du produit ou de l’emballage ; produit ne respectant pas la température acceptable ; date de livraison en désaccord avec la commande ; livraison en dehors de l’horaire de livraison erreur d’expédition (volume ou produit incorrect, date de validité dépassée) ; adresse de livraison non localisée ; véhicule sinistré ;

MOTIFS DU RETOUR Commercial produit en non conformité avec la commande (prix, quantité, date de livraison, date de validité) ; commande annulée ; produit ayant atteint sa fin de vie utile; produit saisonnier; conditions commerciales différentes de l’accord passé; établissement fermé; problème de capacité de stockage du client; données incorrectes sur le reçu fiscal ; sur définition de l’accord commercial, comme l’excès de stock dans le canal.

Qualité non-conformité entre emballage et produit; contamination microbiologique ou chimique; produit comportant des erreurs dans le processus de fabrication ou d’emballage; problèmes avec les caractéristiques sensorielles des produits : apparence, couleur, odeur, quantité de glace ou eau issue du dégel, texture.

Tableau 5 – Motifs de retour dans les entreprises étudiées au Brésil Source : élaboré par les auteurs Le rappel est un type de retour effectué lorsqu’il est détecté un problème de qualité intrinsèque du produit ou une contamination. Les entreprises ont affirmé que ce retour est assez rare, mais crucial, puisqu’il implique la garantie de la santé du consommateur et l’image de l’entreprise. Ainsi, toutes les entreprises ont affirmé posséder une planification et une formalisation des étapes à suivre. Le processus de rappel utilisé par le distributeur F, entreprise qui a fourni le plus de détail sur la procédure, est illustré dans la Figure 5. Identification du problème Analyse du produit Contact avec le fabricant

Le fabricant ne se responsabilise pas pour le rappel

Le fabricant reconnaît sa responsabilité dans le problème

Produit non vendu au consommateur

Produit vendu au consommateur

Retrait des produits des magasins et consolidation de la charge pour collecte du fabricant

14

Communication aux consommateurs et échange ou remboursement des produits

Retrait du produit des magasins et stockage Début du procès juridique contre le fabriquant

Mise au rebut des produits après expertise


Figure 5 – Processus de rappel utilisé par le distributeur F Source : élaboré par les auteurs Les processus les plus souvent utilisés par les entreprises de la grande distribution sont la consolidation des charges pour la collecte par l’industriel et la mise au rebut des produits (par incinération ou enfouissement). Les entreprises n’effectuent pas de donation de produits périssables d’origine carnée ou lactée en raison d’un risque de rupture de la chaîne complexe de froid qui pourrait nuire à la santé du consommateur et par conséquent à l’image de l’entreprise. Quant aux facteurs critiques gérés par la logistique inversée, il a pu être observé que les contrôles d’entrées et les relations collaboratives de la chaîne sont les deux éléments les plus importants. En plus des criticités évoquées par la littérature, la question fiscale associée aux flux des retours est aussi cruciale. Certains interviewés affirment que la législation est d’une extrême complexité et la moindre erreur sur un document fiscal peut impliquer une annulation du crédit d’impôt. Cette réglementation empêche également toute pratique de donation. Un autre point important est l’élaboration des contrats qui agit sur la minimisation des retours en définissant la répartition des responsabilités. Les entreprises étudiées qualifient les relations entre les industriels et les distributeurs du canal de distribution de produits alimentaires d’origine carnée et lactée de normales, malgré le sentiment unanime des industriels selon lesquels la grande distribution utilise son poids pour leur imposer la gestion de la logistique inversée. La forme de pouvoir la plus citée est la coercition, où un membre est puni s’il n’agit pas selon la volonté de l’autre. Selon les interviewés, l’intensité des conflits ne modifie pas le caractère collaboratif des relations. Le pourcentage de retour de produits est inférieur à 5% dans les entreprises étudiées, mais cette valeur a été divulguée uniquement par les industriels. La performance de la logistique inversée est comparable à celle de la logistique directe pour les entreprises A, D et G, mais pire selon l’entreprise C et bien pire selon les distributeurs E et F. Par ailleurs, seules les sociétés D et G ont atteint les résultats escomptés pour la gestion des flux inversés dans le canal de distribution alimentaire. Il a été demandé aux interviewés d’identifier dans un tableau les mesures que l’entreprise utilise pour évaluer la logistique inversée. De plus, ils ont été sollicités pour préciser les

15


mesures qu’ils estiment importantes, même si l’indicateur n’est pas en place dans l’entreprise pour cette activité. Les résultats sont regroupés dans le Tableau 6, les chiffres représentant le nombre de citations des mesures. Certaines mesures de performance sont utilisées par les sept entreprises étudiées : le coût total de la logistique inversée, le coût des marchandises retournées, le coût des produits avariés, le coût des retours des emballages et des palettes, la précision des commandes envoyées, le pourcentage des retours et les motifs des retours. D’autre part, les mesures considérées comme les plus révélatrices ont été la traçabilité/sécurité alimentaire, le nombre de commandes à problème, le coût administratif de la logistique inversée, le délai de résolution des problèmes et le retard moyen. Il existe peu de mesures relatives aux pertes des produits dans les magasins du distributeur car elles sont la plupart du temps comprises dans le coût des produits invendus. TYPE

Coûts

Gestio n des actifs

Service au Client

MESURE Coût total de la logistique inversée Coût des retours en relation avec le coût total de la logistique Coût des retours par type de produit et de fournisseur Coût des marchandises retournées Coût des produits invendus par le distributeur Coût des produits avariés Coût des erreurs du prestataire de services Coût des retours d’emballages et de palettes Coût de résolution des litiges Coûts administratifs de la logistique inversée Coût de la négociation relative aux retours Coûts des invendus à l’origine d’une erreur de commande Niveau de stock Rotation du stock Obsolescence Précision des commandes passées Temps de lancement d’une commande Ponctualité (nombre de commandes livrées à temps) Retard moyen % de demande de retours non prévus au contrat mais acceptés par le fournisseur Nombre de commandes à problèmes % de commandes à l’origine de réclamation % de retour des produits % d’échange des produits % de retour des produits par client / magasin % d’échange des produits par client / magasin Action engagée pour la résolution du problème Motifs des réclamations Délai de résolution des problèmes Information anticipée d’annulation ou de retard Traçabilité, Sécurité alimentaire % de matériels susceptibles d’être recyclés % d’articles incorrects sur une commande Envoi d’une commande à une adresse erronée Intégrité de la marchandise

16

IMPLANTE E 7 1 7 1 7 1 7 2 1 1 2 2 4 7 1 5 1

RELEVATRIC E 4 3 1 3 2 5 1 4 2 1 1 4 1 5

1 1 7 3 1 2 7 2 1 3 6 4 1

5 1 2 3 3 5 4 6 2 2


Tableau 6 – Mesures de performance de la logistique inversée utilisées (ou révélatrices) par les entreprises étudiées au Brésil Source : élaboré par les auteurs Dans les dimensions de la gestion des actifs et du service au client, les mesures présentées dans la littérature et celles rencontrées dans la recherche de terrain sont assez similaires. Dans la dimension du service au client, trois mesures - information anticipée d’annulation ou de retard, % d’échange des produits par client / magasin, % de commandes à l’origine de réclamation – ne sont utilisées par aucune entreprise. Parmi celles-ci, la mesure information anticipée d’annulation ou de retard a été considérée comme rélévatrice par quatre des sept entreprises consultées. Dans la dimension des coûts, le coût de la négociation relative aux retours, la rentabilité de la logistique inversée et le coût de la prévention des retours ne sont utilisés par aucune entreprise étudiée. Par ailleurs, la recherche empirique a fournie certains indicateurs utilisés par les entreprises et non cités dans la littérature tels que : le coût des retours par type de produit et de fournisseur, le coût des produits invendus, les coûts administratifs de la logistique inversée, le coûts des invendus à l’origine d’une erreur de commande. L’analyse des résultats montre que la logistique inversée est une activité qui gagne sa place dans les entreprises dont certaines s’engagent dans l’amélioration des pratiques et des processus. Ces changements impliquent parfois des modifications dans la gestion de l’activité ainsi que des restructurations organisationnelles, comme chez les distributeurs F et G. Chez le distributeur F, la logistique inversée a gagné de l’importance suite à des problèmes répétés relatifs à la qualité des produits, aux emballages et à des dates de validité occasionnant de fortes pertes financières. Chez le distributeur G, des changements ont été opérés depuis 2006 afin de promouvoir les améliorations de la logistique inversée. Au cours de ces modifications, la société à créé une division spéciale pour la logistique inversée afin d’administrer les retours et les rebuts après le transfert de propriété des produits. Cette division est subordonnée à la logistique intégrée et fait partie du département de gestion de la chaîne d’approvisionnement. Le service de gestion des stocks vérifie et contrôle l’entrée des produits dans le centre de distribution et est chargé des flux inversés sans transfert de propriété. Malgré tous les changements et les améliorations constatés, l’analyse des résultats permet d’affirmer que la logistique inversée est plus élaborée chez les industriels que chez les distributeurs. Selon Lavalle et Fleury (2000), les entreprises industrielles se caractérisent par des processus logistiques et productifs d’une plus grande complexité que celle de la 17


distribution. Les industriels ont développé de meilleurs niveaux de sophistication de leur organisation logistique que les distributeurs avec pour objectif une plus grande flexibilité pour devenir plus compétitifs. Il est tout de même notable que les distributeurs ont été les moteurs de ces changements, en exigeant continuellement des industriels un meilleur taux de service et une performance opérationnelle accrue. Dans les trois réseaux de grande distribution étudiés, le distributeur G se dégage par le développement de sa logistique inversée. Cette disparité entre le distributeur G et les distributeurs E et F a pour origine leur attitude quant à la stratégie adoptée vis-à-vis de la logistique inversée : le distributeur G la considère comme essentielle pour la gestion de ses stocks. Par contre, les distributeurs E et F ne considèrent pas cette activité comme stratégique, même s’ils pensent qu’elle est importante pour la gestion des flux inversés, sans pourtant lui conférer des avantages compétitifs notables. C’est pourquoi il est possible d’affirmer que les distributeurs E et F sont encore à un stade de développement initial de la logistique inversée. Du côté des entreprises productrices, les résultats ne permettent pas d’affirmer qu’il y ait une gestion de la logistique inversée plus évoluée chez l’une que chez les deux autres. Les entreprises ont énoncé certaines spécificités relatives au marché des produits carnés et lactés. La principale différence impactant la logistique inversée entre ces deux produits vient de la durée de vie inférieure des produits lactés. Ainsi, ces deux types de produits possèdent des politiques commerciales différentes qui interfèrent avec la logistique inversée.Le rapport des compensations financières payées aux distributeurs pour l’indemnisation des pertes est supérieur pour les produits lactés. Même dans la catégorie lactée, cette valeur varie en fonction de la périssabilité du produit (plus de valeur pour les yaourts que pour les fromages, par exemple) et du client. D’un autre côté, les pertes de produits carnés sont plus faibles, le niveau de performance et de conscience des fournisseurs étant plus importantes au regard de la qualité des produits. Le marché carné est plus concentré et les grandes entreprises exportent également leurs produits sur les marchés internationaux où la qualité / traçabilité / sécurité des aliments sont primordiales. Cette maîtrise confirme l’opinion des distributeurs qui affirment avoir moins de problème avec les produits d’origine carnée. 5. Conclusion Cette recherche présente des limites, la première est l’impossibilité de comparer les résultats de la recherche réalisée en France avec celle effectuée au Brésil. L’objectif des deux recherches était différent, ceci limite également l’interprétation des résultats. Pour des recherches futures liées au sujet, il serait judicieux de réaliser une étude permettant la 18


comparaison avec les entreprises françaises du canal de distribution ou une étude comparative avec les canaux de distribution d’autres pays comme les Etats-Unis et l’Angleterre, pays se caractérisant par leur industrie alimentaire et leur mode de distribution évolués. L’autre limite de cette étude est la difficulté à généraliser les résultats obtenus sur des marchés et des entreprises différentes en élargissant les recherches à d’autres canaux de distribution. L’étude a pourtant atteint son objectif d’une meilleure compréhension des pratiques de la logistique inversée et des éléments qui doivent être gérés pour obtenir une bonne performance de l’activité dans un secteur spécifique de l’économie, dont la logistique est considérée comme complexe. Les interviewés se sont accordés pour dire que la périssabilité des produits exige un grand contrôle sur les processus et sont plus influencés par les décisions commerciales comme les conditions de vente, les compensations et les remboursements, que les produits de mercerie classique. En ce sens, les résultats montrent quelques mesures pour minimiser les retours comme l’attention à l’élaboration des contrats (avec l’identification des responsabilités et la rédaction de procédures détaillées par exemple), la formalisation de la logistique inversée, une mise en place efficace des pratiques commerciales de négociation, l’introduction de la logistique inversée comme un élément du niveau de service, une meilleure prévision de ventes / d’achat et un travail en partenariat des entreprises avec leurs clients. L’analyse des résultats montre que la logistique inversée est une activité qui gagne une place importante dans les entreprises. Pourtant, les résultats montrent également qu’il existe encore des entreprises qui ne considèrent pas la logistique inversée comme stratégique, principalement dans la grande distribution. Tant que les entreprises n’appréhenderont pas la logistique inversée comme une activité capable de générer des avantages compétitifs, toute la structure, l’opérationnalisation et la performance de cette activité seront compromises. Lorsqu’ un tel désintérêt est porté à cette activité, la Direction Générale néglige la logistique inversée et peu de ressources lui sont attribuées. En conséquence, le manque de ressources humaines et de systèmes adéquats limite son implantation et son développement. La faible importance attribuée à la logistique inversée par rapport à d’autres fonctions de l’entreprise est l’obstacle majeur à son développement. Son implantation, son expansion ou son amélioration sont compromises par les efforts engagés dans d’autres services plus cruciaux ou prioritaires du point de vue de l’entreprise. Il n’est pas affirmé dans cette étude que la logistique inversée est plus importante que les autres opérations, mais elle est sans aucun doute une activité essentielle dans la gestion des flux. Sa performance est compromise

19


par le manque d’attention qui lui est portée, ce qui, par conséquent, réduit le potentiel d’une activité génératrice d’avantages compétitifs pour les entreprises. Il convient toutefois de rappeler l’importance d’une logistique de distribution efficiente pour améliorer la logistique inversée. Au Brésil, la logistique n’a pas encore atteint sa maturité et les entreprises concentrent leurs ressources sur son optimisation. Lorsque les problèmes et l’inefficience actuelle de la logistique seront surmontés, les efforts des Directions Générales seront consacrés à l’amélioration d’autres activités, dont la logistique inversée.

Bibliographie ASSOCIAÇÃO BRASILEIRA DE SUPERMERCADOS (ABRAS). (2008) O setor supermercadista brasileiro: totais do setor. Disponível em: http://www.abras.com.br Acesso em: outubro de 2008. BEAMON, B. M. (1999) Measuring supply chain performance. International Journal of Operations & Production Management, vol. 19, n. 3, p. 275-292. CHAVES, G.L.D. (2009) Logística reversa de pós-venda para alimentos derivados de carnes e leite: análise dos retornos de distribuição. Tese (doutorado) Universidade Federal de São Carlos, São Carlos, 2009. 302p. COUNCIL OF SUPPLY CHAIN MANAGEMENT PROFESSIONALS (CSCMP). (2005) Supply chain and logistics terms and glossary, 2005. Disponível em: http://www.cscmp.org/Terms/glossary03.htm Acesso em: janeiro de 2005. DE BRITO, M.P. (2004) Managing reverse logistics or reversing logistics management? 2004. 324p. Tese (doutorado) - Erasmus University Rotterdam, Rotterdam. DEKKER, R. et al. (2004) Reverse logistics: quantitative models for closed-loop supply chains. Berlin: Springer-Verlag. ECR BRASIL. Diferencial que ninguém ensina. Boletim da Associação ECR Brasil sobre eficiência empresarial, ano 2, n. 5, outubro 2002. Disponível em: http://www.ecrbrasil.com.br/ecrbrasil/page/noticias_ecr.asp Acesso em: novembro de 2008. FLEISCHMANN, M. (2001) Quantitative models for reverse logistics. Berlim: Springer. GLOBAL RESEARCH TEAM AT MICHIGAN STATE UNIVERSITY. (1995) World class logistics: the challenge of managing continuous change. Oak Books, IL, EUA, Council of Logistics Management, 423 p. HIJJAR, M. F. et al. (2005) Mensuração de desempenho logístico e o modelo World Class Logistics (Partes 1 e 2). 2005. Disponível em: http://www.cel.coppead.ufrj.br/fsbusca.htm?fr-public.htm Acesso em: janeiro de 2006. LACERDA, L. (2003) Logística Reversa: Uma visão sobre os conceitos e as práticas operacionais. In: FIGUEIREDO, K. et al (orgs.) Logística e gerenciamento da cadeia de suprimentos: planejamento do fluxo de produtos e dos recursos. Centro de Estudos em Logística. COPPEAD, UFRJ. São Paulo: Atlas. LAMBERT, D.M et al. (1998) Supply chain management: implementation issues and research opportunities. The International Journal of Logistics Management, vol. 9, n. 2, p. 1-19. LAMBERT, D.M; POHLEN, T.L. (2002) Mesurer la performance globale de la chaîne logistique. Logistique & Management, vol. 10, n. 1, p. 3-20. LAVALLE, C.; FLEURY, P.F. (2000) Avaliação da Organização Logística em Empresas da Cadeia de Suprimento de Alimentos: indústria e comércio. Revista de Administração Contemporânea, vol. 4, n. 1, p. 47-67, Jan./Abr. 20


LEITE, P R. (2003) Logística reversa: meio ambiente e competitividade. São Paulo: Prenctice Hall. LEITE, P.R.; BRITO, E.P.Z.; SILVA, A.A. (2008) Hábitos empresariais brasileiros em logística reversa. In: XI Simpósio de Administração da Produção, Logística e Operações Internacionais – XI SIMPOI, São Paulo, 2008. Anais... São Paulo, FGV. REVISTA EXAME. (2008) Melhores e maiores 2008. Edição 0922A julho 2008. Ed Abril. ROGERS, D.S.; TIBBEN-LEMBKE, R.S (1998) Going backwards: reverse logistics trends and practices, University of Nevada. Reno: CLM, 1998.283p. ______. (2001) An examination of reverse logistics practices. Journal of Business Logistics, vol. 22, n. 2, p. 129-148. STANK, T.; CRUM, M.; ARANGO, M. (1999) Benefits of interfirm coordination in food industry supply chains. Journal of Business Logistics, vol. 20, n. 2, p.21-41. SUPERHIPER. (2008) Panorama 2008. Revista SuperHiper, ABRAS, maio de 2008. THE EUROPEAN WORKING GROUP ON REVERSE LOGISTICS (REVLOG). (2005) Disponível em: http://www.fbk.eur.nl/OZ/REVLOG/Introduction.htm. Acesso em janeiro de 2005.

21



Developing capabilities to manage offshore operations: an exploratory study Vilmar Antonio Gonçalves Tondoloa Universidade do Vale do Rio dos Sinos, School of Management and Economic Sciences, Brazil [email protected]

a,b,c

Hale Kaynak College of Business Administration, The University of Texas – Pan American, Texas, USA [email protected] Cláudia Cristina Bitencourtb [email protected] Yeda Swirski de Souzac [email protected]

Abstract To explore how companies develop capabilities for managing offshore operations, this study examines the offshore operations of three companies in Brazil, Denmark, and Germany. The research results provide valuable information on the types of offshore operations implemented by the companies, the strategic role of offshore operations, barriers to implementing offshore operations, capabilities developed by companies for implementing offshore operations, and the role of dynamic capabilities elements (paths, positions, and processes) in the development of capabilities for managing offshore operations.

Key words: Offshoring, Capabilities, Dynamic Capabilities; Operations, Development

1


1: INTRODUCTION Offshoring is a growing operations practice worldwide. Over the last decade, companies have moved manufacturing operations abroad, primarily from developed to developing countries. This movement can be considered a strategy formulated in response to the increasing competitiveness of global markets. In recent years, companies have also moved services, high-skill, and core business activities overseas. This shift in offshoring to more complex operations may require the creation and implementation of new organizational practices that have implications for various organizational issues (Duke CIBER/Archstone Consulting, 2005, 2006; Duke CIBER/Booz Allen Hamilton Inc., 2007). Although offshoring has been practiced by companies for a long time (Hagell III & Brown, 2005; Lewin & Peeters, 2006a; Niederman, 2005; Olsen, 2006; Stringfellow et al., 2007; Sturgeon & Florida, 2000) academic efforts are needed to achieve a full understanding of this phenomenon. One of these efforts is to acquire a better understanding of the strategic aspects of offshoring. Offshoring has implications for the strategic management field because it may require new resources and the development of new or unfamiliar capabilities. Thus, offshoring can be considered an internal process as well as a business strategy for effective management of resources and firm-level capabilities (Doh, 2005). It is also a strategy conducted on a learning-by-doing basis. This aspect suggests that implementation of offshoring is achieved by a continuum of stages (Lewin & Peeters, 2006a; Marskell et al., 2006), during which the development of capabilities makes an important contribution to the implementation and management of this process (Carmel & Agarwal, 2002; Ellran et al., 2008; Levy, 2005; Lewin & Peeters, 2006b; Venkatraman, 2004; Youngdahl & Ramaswamy, 2008). The literature has suggested that capabilities development is important when undertaking more complex offshoring processes such as product development (Manning et al., 2008), and overcoming difficulties created by temporal and spatial distance between locally dispersed work teams (Levina, 2007; Levina & Vaast, 2008). Focusing on managerial and firm capabilities, the dynamic capabilities (DC) approach can be a useful perspective for examining how companies develop unique capabilities in offshoring (Doh, 2005). This study aims to explore how companies develop capabilities for managing offshore operations. It integrates DC as a main theory lens and offshore operations as organizational context. More specifically, this study defines DC as “a firm’s ability to integrate, build, and reconfigure internal and external competencies to address rapidly changing environments” (Teece et al., 1997, p. 516). Such a concept emphasizes DC as a set 2


of organizational processes that result in the development of specific capabilities to fit environmental conditions. To explore how companies develop capabilities for managing offshore operations, this study examines the offshore operations of three companies in Brazil, Denmark, and Germany. The qualitative data were collected through semi-structured interviews with four managers of those companies. This study has a twofold contribution to research. First, it explores how companies develop capabilities for managing offshore operations. Second, it explores the role of three DC elements (paths, positions, and processes), to be discussed later, on the development of such capabilities. This study also contributes to practice by providing insights into how companies have developed capabilities and which capabilities they have developed for managing offshore operations. This article is organized as follows. In the next section we present a theoretical background on offshore operations and DC. In section 3 we describe the methodological procedure adopted. In section 4 we present cases and cross-cases analysis. The paper concludes with discussion of the results as well as implications for researchers and managers. 2: THEORETICAL BACKGROUND 2.1: Offshore operations Contrary to what is commonly believed, offshoring has been practiced by companies for a long time. Ford Motor Company started to produce abroad in 1904, and in Europe, German Daimler started to produce abroad in 1891 (Sturgeon & Florida 2000). Thus, offshoring actually is not a new phenomenon (e.g. offshore plants, Moxon 1975). In the literature, however, the term “outsourcing” is sometimes inappropriately used for “offshore,” and the types of offshore have not been clearly delineated. Outsourcing is a contractual agreement between a company and an external provider to obtain goods and/or services (De Vita & Wang 2006). Offshore, however, can be defined as “the movement or relocation of domestic firm activities and operations abroad” (Bunyaratavej et al., 2008, p. 227). Thus, the main difference is that, in outsourcing, an external provider can be either in the same country or abroad, whereas offshore or offshore sourcing implies that the external provider is located abroad (Chakrabarty 2006a; Niederman 2005; Terjessen 2006). The literature review reveals three types of offshoring: offshore outsourcing, offshore partnership and offshore captive (Jahns et al., 2006; Robinson & Kalakota, 2004; Youngdahl et al., 2008). In this study, we define offshore outsourcing as a company’s transfer of tasks or business functions (e.g. assembly) to a third party in a foreign country (e.g. Beugré & Acar 2008; UNCTAD, 2004). We define offshore captive as a company’s shifting tasks or business 3


functions to its own facilities in a foreign country (e.g. Beugré & Acar 2008; UNCTAD, 2004). We define offshore partnership as an interorganizational relationship that shares tasks or business functions (e.g. joint ventures) in a foreign country (e.g. Robinson & Kalakota, 2004; Youngdhal et al., 2008). As a managerial process, the spread of offshoring is due to the development of organizational and managerial capabilities to coordinate this process (Levy, 2005). Offshoring may be characterized as a learning-by-doing process evolving from experimental practicebased peripheral activities to core business activities. As noted earlier, this aspect suggests that implementation of offshoring is done in a continuum of stages based on learning and capability building (Lewin & Peeters, 2006b; Maskell et al., 2006). Accumulated experience also contributes toward high-skill offshoring activities (Hagel III, 2004). The potential for achieving positive results of offshoring also depends on how companies carry out this process. Consequently, at more advanced stages of offshore operations companies must develop specific capabilities to manage offshore relationships and global networks (Askin & Massini, 2008; Carmel & Agarwal, 2002; Lewin & Peeters, 2006b; Levina, 2007; Levy, 2005; Venkatraman, 2004). Capabilities development (e.g. coordination of globally dispersed activities) has allowed companies to employ more complex offshoring processes. It involves learning processes and identification of and adaptation to changing requirements of and opportunities for offshoring (Manning et al., 2008). 2.2: Dynamic capabilities In the literature on strategy, Makadok (2001) highlights two approaches: resourcepicking and capability-building, two approaches that aim at understanding how the managers generate economic rents for their businesses. The first approach is linked to the ResourceBased View (RBV), which says that businesses achieve better performance in relation to competitors through different resources. The second approach is linked to DC. The DC perspective suggests that firms achieve performance superior to their competitors through development of resources and capabilities (Makadok, 2001). The main issue in this second approach is the relationship between the development of new capabilities and organizational performance (Sapienza et al., 2006). In other words, the DC perspective extends the RBV argument by addressing how resources and capabilities can be created and how the current stock of resources and capabilities can be refreshed in changing environments (Ambrosini & Bowman, 2009). The DC approach includes an evolutionary vision, “Schumpeterian,” of the competition among firms. For this reason, differences among firms are created by the new 4


combinations of resources and capabilities developed by firms throughout their path (Teece et al., 1997). The DC perspective emphasizes two main elements: the dynamic and the capability. The term “dynamic” refers to the shifting character of the environment that requires strategic responses (e.g. renew competences), and the term “capability” refers to the role of strategic management in coping with changing environmental requirements by adapting the company internally (e.g. integrating, and reconfiguring internal and external organizational skills, resources, and functional competences) (Teece & Pisano, 1994; Teece et al., 1997). For this reason, the main concern is the ability of the organization to develop highlevel capabilities through its path to leverage and/or sustain superior performance (Helfat & Peteraf, 2003; Marcus & Anderson, 2006). From the DC perspective, development of resources and capabilities is internal to a firm. In this respect, Eisenhardt and Martin (2000) emphasize that internal processes are the sources of dynamic capabilities. The path of the organization leads to the accumulation of knowledge that is capable of generating new routines and processes (Sapienza et al., 2006). As discussed above, DC can be considered a dynamic perspective of a strategic resource approach. This perspective was introduced by Teece et al. (1997). Since then, other definitions and concepts of DC have been offered in the literature. An analysis of these concepts makes it possible to point out the following aspects. First, the main result of DC is to allow firms to keep up with environmental changes by creating, renewing or integrating resources, assets, capabilities, competences, and routines. As Wang and Ahmed (2007, p. 40) argue, “capability development as an outcome of dynamic capabilities over time is frequently discussed and evidenced in empirical research.” Second, different aspects of DC are emphasized: organizational process, behavioural orientation, high-level managerial activities, firm’s ability, firm’s capacity, firm’s routines and processes, learning process, firm’s activities, and patterns. Thus, it is impossible to argue that DC is based on just one or a few aspects of a firm. In others words, DC is considered a set of organizational aspects that allow companies to cope with new external and internal requirements. This study utilizes a DC approach based on the studies by Teece et al. (1997) and Teece and Pisano (2004) (see Figure 1). The three specific aspects of DC are elements, firmspecific processes, and outcomes. These are common features; in other words, any company should present these aspects embedded in DC (Wang & Ahmed, 2007). The three DC elements help determine a company’s DC and distinctive competence as follow: (1) organizational processes, which entail the organizational and managerial routines of current practice and learning, (2) positions, which refers to a company’s current endowment of 5


technology and intellectual property and its relationships with customers, suppliers, and strategic alliances, and (3) paths, which refers to the strategic alternatives and opportunities available to the company. Figure 1: A framework for developing capabilities to manage offshore operations

Offshoring Elements • Type of offshore operations • Strategic role of offshore operations • Barriers to

implement/manage offshore operations

DC Elements • Organizational processes • Paths • Positions

Firm Specific DC Processes • Reconfiguration • Leveraging • Learning

Outcomes • Developed capabilities to overcome barriers

Dynamic Capabilities

In this study, firm-specific processes of DC entail reconfiguration, leveraging, and learning. These processes may vary among companies because they are developed over time (Wang & Ahmed, 2007). In other words, they are path dependent. Reconfiguration is the recombination of resources and capabilities to fit with changing requirements (Ambrosini & Bowman, 2009; Bowman & Ambrosini, 2003; Eisenhardt & Martin, 2000; Menon, 2008). Leveraging is the replication of a process or systems to another business unit (Ambrosini & Bowman, 2009; Bowman & Ambrosini, 2003). Learning is the creation and regeneration of new knowledge that allows a task to be performed (Ambrosini & Bowman, 2009; Bowman & Ambrosini, 2003; Menon, 2008). In this study, the outcome of DC is the development of capabilities for managing offshore operations. In other words, we refer to capability development as an outcome of a firm’s dynamic capabilities over time. DC can be considered a higher-order capability (e.g. Collis & Montgomery, 1994) or a dynamic ability (e.g. Zahra et al., 2006) that develops, reconfigures, renews, and integrates a company’s capabilities. Therefore, this work focuses on DC as a set of processes for developing organizational capabilities (e.g. the capability for managing offshore operations). In doing so, this work does not discuss DC as a specific distinctive capability (e.g. R&D). 3: METHODOLOGY Because this study is exploratory, we employed the multiple case study research approach. Several studies have discussed the benefits and characteristics of this methodology (e.g. Eisenhardt, 1989; Meredith, 1998; Voss et al., 2002; Yin, 2003). By using multiple cases, for example, we can utilize our theory to highlight certain characteristics, and then find the differences and similarities among the cases (Yin, 2003). To ensure quality of data collection, three companies were selected. The main selection criterion was that companies 6


have had offshore operations. The companies’ headquarters are located in Brazil, Denmark, and Germany. In order to protect the identity of each company, we will refer to these companies as company A, company B, and company C, respectively. The data were collected through semi-structured interviews of executives who have had experience with offshore operations These interviews were conducted by phone and lasted from 50 to 70 minutes. All interviews utilized a protocol. This instrument was developed to ensure valid data was obtained, formalize the case study, and systemize the observation and analysis to increase reliability of the study (internal validity). The use of multiple sources enabled us to triangulate data as a means of validating and verifying the consistency of the data (Stake, 1998; Yin, 2003). To establish external validity, we used analytical generalization based on Yin’s suggestions (2003). To analyze the data we used the qualitative content analysis technique (Flick, 2002; Cooper & Schindler, 2003). The main categories of analysis developed were the following: (1) strategic role of offshore operations, (2) barriers to the implementation of offshore operations, (3) capabilities developed by companies to manage and implement offshore operations, and (4) the role of dynamic capabilities elements (paths, positions, and processes) on the development of capabilities for managing offshore operations. 4: CASES 4.1: Company A Company A is a Brazilian firm established in 1949 that produces a diverse line of coaches used for public transportation. Company A has more than 12,000 employees and operations in multiple countries in addition to Brazil: Argentina, Colombia, Mexico, India, Russia, Egypt, South Africa, and Portugal. It started offshore operations in 1990 with the establishment of a captive factory in Europe. Since then, company A has been moving its operations abroad through captive facilities, acquisitions, joint ventures, and contracts with third party companies. Thus, company A has three types of offshore operations (captive, partnership, and outsource). The managers of Company A perceive differences in the management of these three types of offshoring. According to an interviewee, offshore captive provides the most control over operations and decisions. With the use of other offshoring types, operations management becomes more complex. For instance, in its offshore partnership, company A has to share decisions with its partner. With offshore outsourcing, company A has no control over decisions and operations. Coordination works when using specific contractual agreements in that type of offshore operation. Thus in terms of management and coordination of offshore 7


operations, partnerships and the other types of outsourcing complicate the management of operations for company A. In the early 1990s, company A decided to implement its growth strategy. The company decided to carry out the internationalization process as its main growth strategy. In that way, offshore operations contributed to the implementation of this strategy. Thus, company A highlights the main strategic role of offshore operations: it contributes to the implementation of an internationalization process, and consequently, the effectiveness of the company’s growth strategy. For instance, by 2008 the company’s net income grew from US$ 200 million to more than US$ 1,000 million. Company A sees implementation of its management systems in facilities abroad as the main barrier to implementing offshore operations. Cultural differences, language, and the adaptation of expatriates are other barriers identified by company A. In order to implement offshore operations and overcome such barriers, company A has been providing special training to executives who will be expatriated. In general they are Brazilians with some knowledge of the country in which they will be posted. Company A has also developed its own production system, which is transferred to facilities abroad to ensure standardization of production processes and products. According to an interviewee, company A uses the same operational logic of McDonald’s. Regarding the company’s path, the respondent emphasized that company A has been learning how to deal with cultural differences. This learning was central to the adaptation of expatriates and the implementation of management and production systems. In other words, understanding cultural differences was central to implementation of offshore operations. Finally, the company’s position with regard to its production of a majority of its own components has contributed to the implementation of its offshore operation. Through this position company A has been able to sustain its international expansion for 20 years. 4.2: Company B Company B is a Danish firm established in 1955 that produces hydraulic, electric, and such electronic systems as valves. Company B has more than 5,000 employees and over 20 manufacturing facilities with operations in more than 24 countries. Company B’s offshoring experience started in 1987 through establishment of a joint venture. These days company B usually prefers captive type offshore operations so that it can ensure it controls operations. Except for a facility in China, it does not establish joint ventures. Company B manages its offshore operations with a hierarchical structure organized by product lines, thus different locations can share the same director. However, each facility has local function managers. 8


The main strategic role of offshore operations highlighted by company B is global production flexibility and cost. Offshore operations enable company B to move production of any product or component to where it is most advantageous. According to an interviewee, moving production from one location to another is very dynamic. At any time one location can become more advantageous than another. Thus production lines are moved among global locations frequently. But the main barrier to offshore operations for company B is related to this movement. When production is moved from one location to another, unused components are left at the previous location. This situation causes disagreement between locations because neither facility wants to assume that inventory. To implement offshore operations and support flexibility among locations, company B has acquired knowledge about the customs operations and laws of the countries in which it has facilities. It has one office responsible for learning how customs works in these countries. Frequently an employee visits each country to obtain more specific information. In addition, company B promotes continuous training and human resources development focused on languages and the laws of international commerce. During its path company B has been learning how to transfer businesses to countries that offer more benefits than Denmark. This accumulated knowledge has been essential to company B’s implementation of offshore operations, and it achieves benefits from its global production flexibility. Finally, the company’s position on maintaining captive operations has contributed to implementation of its offshore operations for it ensures that the company’s objectives and goals are being met by the facilities abroad. Plus it avoids conflicts usually created by other options such as partnership and outsourcing. 4.3: Company C Company C is a German company established in 1996 and dedicated to the production of chemicals. It has about 1,000 employees and more than six manufacturing facilities in six countries: Argentina, Brazil, China, France, India, and Italy. Company C has three types of offshore operations, and its executives perceive differences in the management of each type. Managing offshore captive is less complex than the others because it makes quality, standardization, and control of operations as a whole easier; partnership and outsourcing present additional management complexities and risks because these types do not allow direct control by company C. Costs and proximity to clients are two main strategic roles highlighted by company C as justification for offshore operations. They allow company C to produce in a location that offers the best margins and lowest costs. Offshore operations also allow company C to 9


emphasize geographical proximity to its clients. These two advantages are central to achieving a better competitive position in the global market. Legislative changes in the country where offshore facilities are located represent the main barrier highlighted by company C. For instance, an interviewee from company C commented about the changes in legislation that recently occurred in China. These changes have elevated costs to companies previously established in that country, sometimes making offshore operations there not economically viable. To deal with these additional costs, Company C has moved some operations from China to France. The main point is that when Company C started a facility in China, the government offered several incentives, but after three years, more rigorous legislation has made operations in that country more costly. To implement its offshore operations, company C developed a management system that ensures integrated communication and information among locations abroad. Its management system also supports exchange of employees among locations to promote this integration. In addition, company C developed routines that ensure the exchange of information and experience. Offshore operations are also supported by an organization culture that has been developed during the company path. Even though company C is only 14 years old, it has accumulated 300 years of experience thanks to the founders of its predecessor’s business. Finally, company C’s position reallocating its production abroad has contributed to the implementation of its offshore operations. In addition, the company has been moving production among locations, its transfer of parts of its production from China to France for example. It, too, is promoting a more integrated global production through offshore operations. 4.4: Cross-case analysis After analyzing each company separately, the cases were compared to identify similarities and differences. The comparisons are presented in Tables 1 and 2 and are briefly explained in the rest of this section. When examining offshore operation experiences of the companies, it seems there is little significant difference among them. However, it is possible to perceive that company B has the most experience with offshore operations. Table 1: Offshoring aspects Aspects Experience on offshore operations Types of offshore operations Perceive difference on

Company A Since 1990

Company B Since 1987

Company C Since 1996

Captive Partnership Outsource Yes

Captive

Captive Partnership Outsource Yes

Not applicable

10

RIRL 2010 - Bordeaux September 30th & October 1st, 2010 managing types of offshore operations Strategic role of offshore operations

Growth strategy through internationalization process Implementation of own management system, cultural difference, language, and adaptation of expatriates

Production flexibility and costs

Costs and proximity of main clients

Level of inventory

Legislation changes

Aspects Capabilities developed to implement offshore operations Organizational processes/routines developed Company’s trajectory (path)

Company A Development of expatriates

Company B Knowledge about operation and law of countries’ customs Training and human resource development

Company C Management system

Organization culture

Positions

Own production of majority of its components Leveraging and learning

Learning how to transfer businesses to more advantageous countries Captive operations

Productions reallocation

Learning

Learning

Barriers to implement offshore operations

Table 2: DC aspects

Firm-specific processes utilized

System of production

Learning how to deal with cultural differences

Information exchange

While companies A and C utilize three types of offshore operations (captive, partnership, and outsourcing), company B prefers to utilize offshore captive exclusively. Its only exception is in China, and that is because of the restrictive legislation in that country. Company A and C perceive differences in the management of these distinctive types of offshore operation. They agree that offshore captive is less complex to manage and coordinate than offshore partnership and offshore outsourcing. In addition, company A and C observed that their influence over and control of decisions and operations decrease in offshore partnership and outsourcing, respectively. This characteristic can become a threat or risk to offshore operations. Additionally, to ensure stability of operations, all three companies prefer to employ their own hierarchical structure to control offshore operations. Regarding the strategic role of offshore operations, it was possible to identify some distinctions in each company, particularly when comparing company A to companies B and C. Company A has been utilizing offshore operations to implement its internationalization process, which is aimed at achieving the objectives of its growth strategy. Thus, offshore operations have been contributing to company A’s presence in several countries besides 11


Brazil. Companies B and C clearly have been relying on offshore operations to achieve cost advantages. Achieving this aim is possible because of the global operation flexibility offshore operations allows. In other words, offshore operations permit companies B and C to move their production to locations that can be more advantageous in terms of costs. Regarding barriers to the implementation of offshore operations, it was possible to identify some distinctions in each company. Company A perceives the most diverse numbers of barriers: implementation of its own management system, cultural difference, language, and the adaptation of expatriates. Conversely, companies B and C identify barriers that are focused on a specific aspect of offshore operations. Company B is focused on the barrier related to logistics operations, which highlights possible inventory problems among locations. As Company B constantly moves line production among global locations, it has encountered issues with inventories of components. This situation creates disagreement among facilities because no facility wants to assume a surplus of components. Company C considers legislation changes as the main barrier. These changes are increasing its operational costs. One way or another, Company B and C are both facing barriers that may result in additional costs. Thus, these barriers may be considered examples of costs not readily apparent prior to implementing offshore outsourcing (Ellran et al., 2008). This finding leads us to present the following proposition: P1: Barriers to implementing and managing offshore operations are dynamic throughout the implementation and management of offshore operations. As can be seen in Table 2, all companies perceive the contribution of DC elements and firm specific DC processes to the development of capabilities for managing offshore operations. But each company perceives that contribution differently. Company A considers its development of expatriates as the main resource and capability developed to implement offshore operations. The development of its own executives allows company A to deal with barriers faced in offshore operations. Knowledge is the main resource identified by company B when it comes to implementing offshore operations. That knowledge is focused on supporting the logistics of offshore operations among locations. Thus company B has been developing knowledge on how to operate each location and deal with legislation. This knowledge is seen as essential to ensure agility and flexibility of offshore operations among locations. Company C regards its own management system as the main resource it developed to implement offshore operations. That system allows company C to achieve an integrated communication and information flow among locations abroad.

12


All companies identified contribution of organizational processes to developing capabilities to manage offshore operations; however, they focused on different processes. Company A identified its own system of production as the main process contributing to offshore operation. That system has been implemented in each location to ensure standardization of operations. Company B identified its training and human resource development as its main process. Unlike company A, company B’s process is not focused on development of expatriates, but it is related to the development of employees in each offshore location. This process supports regular operations among locations, particularly those related to production transfer. Company C identified information exchange as the main process it developed. This process allows company C to maintain coordination among locations. On the evidence of the above cross-case analysis and discussion, we present the following proposition: P2: Organizational processes contribute to the development of capabilities for implementing and managing offshore operations; however the capabilities developed may vary among companies. Path also contributes to the management of offshore operations for each company. During its path, company A has been accumulating learning related to dealing with cultural differences. This learning has contributed to implementing and managing offshore operations, particularly regarding the adaptation of expatriates and implementation of its own production system. Company B also considers learning accumulated during its path important. Unlike company A, the learning company B sought was related to the transfer of businesses to countries that offer costs advantages. This learning is essential to its captive offshore operations. Company C attributed to its path the formation of an organizational culture that was fundamental to the implementation and management of offshore operations. On the evidence of the above cross-case analysis and discussion, we offer the following proposition: P3: Development of resources and capabilities for implementing and managing offshore operations is a path-dependence process. Concerning positions, each company has an individual perception of this element. However, all companies consider that their adopted position is contributing to offshore operations. Company A attributes its position on maintaining its own production of the majority of its components the main contributor to the implementation and management of its offshore operations. Company B attributes its position on utilizing offshore captive. In some way, the positions of companies A and B are similar. The two companies have a centralized posture in relation to offshore operations that orients the development of their capabilities. 13


Company C considers its position on production reallocating as the main position contributing to management and development of its offshore operations. On the evidence of the above cross-case analysis and discussion we present the following proposition: P4: Development of capabilities for implementing and managing offshore operations is dependent on a company’s past and current commitment to decisions regarding development of technology and specific assets. Finally, firm-specific processes were also identified as contributors to development of capabilities for managing offshore operations. Company A uses leveraging and learning to replicate its own system of production among locations and make specific adaptations according to site needs. Company B recognizes learning as the main firm-specific process. According to that company, learning is a central process for performing operations at dispersed locations. Company B’s employees are continuously stimulated to revise currents processes so as to generate new knowledge about tasks and routines used at offshore operations. Company C also recognizes learning as a main firm-specific process. Managers of each site meet periodically to share experiences and information. This practice contributes to the sharing of knowledge and learning on the management of offshore operations. It should be noted that the companies did not mention any observation regarding reconfiguration processes. On the evidence of the above cross-case analysis and discussion we present the following proposition: P5: Firm-specific dynamic capabilities contribute to the development of capabilities for implementing and managing offshore operations; however, these dynamic capabilities vary among companies. 5: CONCLUSION The main objective of this study was to understand how companies develop capabilities for managing offshore operations. The findings of this case study research have significant implications for offshore operations and DC. Companies in this study perceive differences in managing the different types of offshore operations. Some studies attribute specific risks and barriers to each type of offshoring (e.g. Ellran et al., 2008; Kumar et al., 2009; Levina, 2007). Our findings indicate that different strategic roles of offshore operations may result in different perceptions of barriers, outcomes, and development of specific capabilities, findings that corroborate the arguments of earlier studies. The strategic role of offshore operations is also related to capabilities development. It is related to the argument that DC processes are driven by company strategy (Wang & Ahmed, 2007). For instance, a company’s decision on type of 14


outsourcing (captive, outsourcing, or partnership) and the degree of offshore operations depends on the role of this process in the company’s strategy (Metters, 2008). All three elements of DC are perceived as contributors to the development of capabilities for implementing and managing offshore operations. This finding supports arguments that capability development is an outcome of dynamic capabilities (Wang & Ahmed, 2007). Our findings also show that the development of capabilities is related to support and management of offshore operations. The literature suggests that the main function of DC is to allow a company to be able to fit continuously (e.g. Ambrosini & Bowman, 2009). Two specific DC processes were observed in this study: learning and leveraging. Indeed, learning was identified by companies as an aspect contributing to implementing and managing offshore operations. Companies in our study are searching for ways to replicate and disseminate processes from one location to another to support the implementation and management of offshore operations. In sum, there are common features on DC (elements) and firm-specific processes (DC processes). The common features inherent to DC and firmspecific processes may be dependent on the company’s strategy (Wang & Ahmed, 2007). Company A focuses on its own production system, company B focuses on training and human resource development, and company C focuses on information exchange among locations. This study contributes to practice by providing information on what kind of capabilities companies have been developing, and how they are developing these capabilities for managing offshore operations. Practitioners should first identify the strategic contribution of offshore operations to align it with the company’s operations. It is also important to note that this study suggests a company’s strategy as the main driver of DC and consequently the development of capabilities. One of the limitations of this study is its exploratory nature and the small number of cases. Thus we recommend complementing this study by examining a larger number of companies that have offshore operations. This research is one of the first studies to look at how companies develop capabilities for managing offshore operations. Future research studies on DC and offshore operations in other companies that engage in offshore operations can continue contributing to theory building and testing in this area.

6: REFERENCES Ambrosini, V., Bowman, C. (2009), What are Dynamic Capabilities and are They a useful Construct in Strategic Management?, International Journal of Management Reviews, Vol. 11, No 1, pp. 29–49.

15


Askin O. Z., Masini, A. (2008), Effective Strategies for Internal Outsourcing and Offshoring of Business Services: An Empirical Investigation, Journal of Operations Management, Vol. 26, pp. 239–256. Beugré C. D., Acar W. (2008), Offshoring and Cross-Border Interorganizational Relationships: A Justice Model, Decision Sciences, Vol. 39, pp. 445–468. Bowman, C., Ambrosini, V. (2003), How the Resource-Based and the Dynamic Capability Views of the Firm Inform Corporate-Level Strategy, British Journal of Management, Vol. 14, pp. 289–303. Bunyaratavej, K., et al. (2008). Multinational investment and host country development: Location efficiencies for services offshoring, Journal of World Business, Vol. 43, pp. 227– 242. Carmel, E., Agarwal, R. (2002), The Maturation of Offshore Sourcing of Information Technology Work, MIS Quarterly Executive, Vol. 1, No 2, pp. 65–78. Chakrabarty, S. (2006), Making Sense of the Sourcing and Shoring Maze: Various Outsourcing and Offshoring Alternatives, in Kehal, H. and Singh, V. P. (Eds.), Outsourcing and offshoring in the 21st Century: a socio economic perspective, Idea Group Publising, Hershey, PA, pp. 18–53. Cooper, D. R., Schindler, P. S. (2003), Business Research Methods (8th ed). McGraw-Hill: New York. Collis, D. J., Montgomery, C. A. (1994), Competing on Resources. Harvard Business Review, Vol. 37, No 4, pp. 118–128. De Vita, G., Wang, C. L. (2006), Developing of outsourcing theory and practice: a taxonomy of outsourcing generations, in Kehal, H. and Singh, V. P. (Eds.), Outsourcing and offshoring in the 21st Century: a socio economic perspective, Idea Group Publising, Hershey, PA, pp. 1– 17. Doh, J. P. (2005), Offshore Outsourcing: Implications for International Business and Strategic Management Theory and Practice. Journal of Management Studies, Vol. 42, No 3, pp. 695– 704. Duke University CIBER/Archstone Consulting (2005), First Biannual Offshoring Survey Results, Avaliable at: https://offshoring.fuqua.duke.edu/pdfs/offshore_duke_particpant_brochure_04-0705_final.pdf (accessed 3 March 2009).

16


Duke University CIBER/Archstone Consulting (2006), Second Biannual Offshoring Survey Results, Avaliable at: https://offshoring.fuqua.duke.edu/pdfs/2nd_highlights.pdf (accessed 3 March 2009). Duke University CIBER/ Booz Allen Hamilton (2007), Next Generation Offshoring – The Globalization of Innovation, Available at: https://offshoring.fuqua.duke.edu/research.jsp (accessed 3 March 2009). Eisenhardt, K. M. (1989), Building Theories from Case Study Research, The Academy of Management Review, Vol. 14, pp. 532–550. Eisenhardt, K. M., Martin, J. A. (2000), Dynamic Capabilities: What are They?, Strategic Management Journal, Vol. 21, pp. 1105–1122. Ellran, L. M. et al. (2008), Offshore Outsourcing of Professional Services: A Transaction Cost Economics Perspective, Journal of Operations Management, Vol. 26, pp. 148–163. Flick, U. (2002), An Introduction to Qualitative Research, 2nd edn. London, Sage. Hagel III, J. (2004). Offshoring Goes on the Offensive. McKinsey Quarterly, Vol. 2, pp. 83– 91. Hagel III, J., Brown, J. S. (2005), The Only Sustainable Edge. Why Business Strategy Depends on Productive Friction and Dynamic Specialization, Harvard Business School Press, Boston, MA. Helfat, C., Peteraf, M. (2003), The Dynamic Resource-Basead View: Capability Lifecycles, Strategic Management Journal, Vol. 24, pp. 997–1010. Jahns, C. et al. (2006), Offshoring: Dimensions and Diffusion of a New Business Concept, Journal of Purchasing and Supply Management, Vol. 12, No 4, pp. 218–231. Kumar, K. et al. (2009), Offshoring and the global Distribution of Work: Implications for Task Interdependence Theory and Practice, Journal of International Business Studies, Vol. 40, pp. 642–667. Levina, N. (2007), Leadership Practices for Achieving Collaboration in Global Sourcing, FSO, March, pp. 27–29. Levina, N., Vaast, E. (2008), Innovating or Doing as Told? Status Differences and Overlapping Boundaries in Offshore Collaboration, MIS Quarterly, Vol. 32, No 2, pp. 307– 332. Levy, D. L. (2005), Offshoring in the New Global Political Economy, Journal of Management Studies, Vol. 42, No 3, pp. 685–693. Lewin, A. Y., Peeters, C. (2006a), Offshoring Work: Business Hype or the Onset of Fundamental Transformation?, Long Range Planning, Vol. 39, No3, pp. 221–239. 17


Lewin, A. Y., Peeters, C. (2006b), The Top-Line Allure of Offshoring, Harvard Business Review, Vol. 84, No 3, pp. 22–24. Makadok, R. (2001), Toward a Synthesis of the Resource-Based and Dynamic-Capability Views of Rent Creation, Strategic Management Journal, Vol. 24, pp. 387–401. Manning, S. et al. (2008), A Dynamic Perspective on Next-Generation Offshoring: The Global Sourcing of Science and Engineering Talent, Academy of Management Perspectives, Vol. 22, pp. 35–54. Marcus, A. A., Anderson, M. H. (2006), A General Dynamic Capability: Does it Propagate Business and Social Competencies in the Retail Food Industry?, Journal of Management Studies, Vol. 43, No 1, pp. 19–43. Maskell, P. et al. (2006), Learning paths to offshore outsourcing: From cost reduction to knowledge seeking, (CBS Center for Strategic Management and Globalization Working Paper 13/ 2006), Copenhagen Business School, Denmark. Menon, A. G. (2008), Revisiting Dynamic Capability, IIMB Management Review, Vol. 20, No 1, pp. 22–33. Meredith, J. (1998), Building Operations Management Theory through Case and Field Research, Journal of Operations Management, Vol. 16, pp. 441–454. Metters, R. (2008), A Case Study of National Culture and Offshoring Services, International Journal of Operations & Production Management, Vol. 28, pp. 727–747. Moxon, R. W. (1975), The Motivation for Investment in Offshore Plants: The Case of the U.S. Electronics Industry, Journal of International Business Studies, Vol. 6, pp. 51–66 Niederman, F. (2005), International Business and MIS Approaches to Multinational Organizational Research: The Cases of Knowledge Transfer and IT Workforce Outsourcing, Journal of International Management, Vol. 11, pp. 187–200. Olsen, K. B. (2006), Productivity impacts of offshoring and outsourcing: A review (OECD STI Working Paper 2006/ 1), OECD Directorate for Science, Technology and Industry, Paris. Robinson, M., Kalakota, R. (2004), Offshore Outsourcing: Business Models, ROI and Best Practices, Mivar Press, Alpharetta, GA (2nd edition). Sapienza, H. J. et al. (2006), A Capabilities Perspective on the Effects of Early Internationalization on Firm Survival and Growth, Academy of Management Review, Vol. 31, pp. 914–933. Stake, R. (1998), Investigación con estudio de casos, Ediciones Morata, Madrid. Stringfellow, A. et al. (2008), Invisible Costs in Offshoring Services Work, Journal of Operations Management, Vol. 26, pp. 167–179. 18


Sturgeon, T. J., Florida, R. (2000), Globalization and Jobs in the Automotive Industry. Services Offshoring Working Group Final Report, Massachusetts Institute of Technology, Boston, MA. Teece, D. J., Pisano, G. (1994), The Dynamic Capabilities of Firms: An Introduction, Industrial and Corporate Change, Vol. 3, pp. 537–556. Teece, D. et al. (1997), Dynamic capabilities and strategic management, in Foss, N. J. (Ed.), Resources firms and strategies – a reader in the resource-based perspective, Oxford University Press, Oxford, pp. 268–285. Terjesen, S. (2006), Outsourcing and offshoring of finance activities, in Kehal, H. and Singh, V. P. (Eds.), Outsourcing and offshoring in the 21st Century: a socio economic perspective, Idea Group Publising, Hershey, PA, pp. 209–228. UNCTAD, (2004), World Investment Report 2004: The Shift Towards Services, United Nations, New York, Available at: http:// www.unctad.org (13 March 2009). Venkatraman, N. (2004), Offshoring without Guilt, MIT Sloan Management Review, Vol. 45, No 3, pp. 14–16. Voss, C. et al. (2002), Case Research in Operations Management, International Journal of Operations & Production Management, Vol. 22, pp. 195–219. Zahra, S. A. et al. (2006), Entrepreneurship and Dynamic Capabilities: A Review, Model and Research Agenda, Journal of Management Studies, Vol. 43, pp. 917–955. Wang, C. L., Ahmed, P. K. (2007), Dynamic Capabilities: A Review and Research Agenda, International Journal of Management Reviews, Vol. 9, No 1, pp. 31–51. Yin, R. K. (2003), Case Study Research: Design and Methods, Sage, Newbury Park, CA (3th edition). Youngdahl, W., Ramaswamy, K. (2008), Offshoring Knowledge and Service Work: A Conceptual Model and Research Agenda, Journal of Operations Management, Vol. 26, pp. 212–221. Youngdahl, W. et al. (2008), Exploring New Research Frontiers in Offshoring Knowledge and Service Processes, Journal of Operations Management, Vol. 26, pp. 141–147.

19



Transformability drivers - a configuration model for modular logistic processes in SME

Prof. Dr- Ing. Frank Straube Chair of logistics, Berlin University of Technology, Germany [email protected] Dipl.-Kffr. Anja Steffens Chair of logistics, Berlin University of Technology, Germany [email protected] Dipl.-Ing. Ouelid Ouyeder Chair of logistics, Berlin University of Technology, Germany [email protected]

Abstract

This article aims to promote science and practice by designing a modularization concept and transferring the experiences to the transformability enablers for a holistic approach. Initially the current state of research is described. A model of coherence of the five transformability enablers will be derived. The second step refers to the configuration potentials the modularization offers to companies. Subsequently it is researched – by applying the case study design – how the transformability driver modularization can be transferred and implemented to the logistics processes of small and medium sized enterprises (SME). Finally those experiences are adapted to the model of transformability enablers.

Key words: Transformability, Modularity, Enabler, Small and Medium sized Enterprises

1


1. INTRODUCTION Innovative solutions contain decisive factors for small and medium sized enterprises (SME) in the production industry. The short communication ways and flat hierarchies inside the company produce a supporting effect, but the restricted personal and financial resources limits the innovation activities (Daschmann, 1994; Ernst, 1999; Bergman & Crespo, 2009; Morgan, 2004). Within a challenging environment, especially during financial and economic crisis the importance of innovative and agile processes rise. The challenge for logistics processes is to manage effective and efficient processes at reasonable costs during intense volume changes. Some of the mainly applied concepts are the development of new business areas, cost reduction by bundling resources and collaboration. But SMEs have difficulties to reduce the complexity of their logistics systems due to their restricted resources (e.g. personnel, finances, skills) (Straube, 2007). However the management of logistics processes has increasingly become a competitive advantage. Particularly, since SME have been operating worldwide in international markets, they need to adapt their organizational and technological structure to the current development (Belliger & Krieger, 2007). Transformability concept comes forward as a solution to restructure a company’s organization. This concept provides five enablers – modularity, mobility, universality, compatibility and scalability. The enablers can be applied in the system and process design to realize a broad variety of actions and reactions to cope with the current market conditions. In the following chapters those enablers will be introduced and described. In order to demonstrate the challenges and complexity of enablers’ implementation in the practice, a concept model of system design based on the modularization which has been developed before in a project for SME is presented. The main research question in this paper is to show how transformability enablers should be designed and deployed to raise the transformability of SMEs. The question will be answered based on three supplement research methods – literature review, empirical study and a conceptual approach. The concept of transformability is presented in the first chapter and different definitions are distinguished from each other and described. An insight in design parameters of transformability is given by analyzing and structuring the enablers. As the enablers are more common in other research fields, different research areas such as production, information technology and economics were probed in order to determine the relevant enablers’ characteristics which can be transferred to logistics processes. The resulting model of

2


enablers clarifies the interdependences between those enablers which are relevant for the design of a holistic approach. The second and third chapters focus on one of the transformability enablers “modularity”. Its characteristics in terms of logistics processes are defined. The results of the second section are mainly based on a literature review in order to convey the idea of logistics modularization for companies in particular for SME. Firstly, an empirical study has been conducted in order to research success factors and requirements of SME when they are going to implement the logistics modularization. The empirical part has been the precondition for the second part which is a conceptual approach. A modularization procedure has been developed based on the previous findings. Finally the findings according to the developed procedure, success factors and requirements are summarized and relayed to transformability concept with its five enablers. The chances and obstacles of an integrated approach including all enablers are discussed and an implementation model is developed.

1.1 Transformability Nowadays companies carry on in a turbulent environment. This requires an extensive agility and reactivity of system behavior (Hernández Morales, 2003). The drivers of change can be reviewed as external and internal change drivers. Intern drivers are for example the location, buildings, staff, organization and management of a company (Tichy, 1983) and can be divided into proactive and reactive change processes (Noffen et al, 2005). Market development, legislation, competition and the development of technologies are extern drivers (Keijzer, 2007; Westkämper & Zahn, 2009). In science different approaches are discussed to deal successfully with those changes and turbulences. Especially the flexibility was discussed broadly and can be defined as an ability to vary inside planned horizon without causing additional costs. (Nyhuis et al, 2008) Besides Hildebrand distinguishes between a basic flexibility and an advanced flexibility, though the advanced flexibility is essentially conform to the definition of transformability (Hildebrand, 2005). Though it cannot be argued that an agreeable and consistent meaning exists in literature (e.g. Shewchuk registered in a literature review over 70 different definition approaches (Shewchuk & Moodle, 1998) and the boarders to transformability are fluent. Transformability is discussed mainly in the German literature since the 90s. (Cisek et al, 2002) The idea of transformability goes beyond the predicted boarders of the basic definition of flexibility - the concept aims to design a system that can realize changes outside estimated ranges very fast and with just small investments. 3


(Hernández Morales, 2003) Such system behavior can be achieved by the implementation of transformability enablers. In particular there is a slight difference between agility and transformability which can vary between different areas of expertise (organization, production etc.). (Möller, 2008) Opposite to the concept of transformability the agility has a broader range of view and refers to the network level. (Goldmann, 1996 and Ramesh & Devadasan, 2007) The concept of agility focuses more on the action and takes the complete system into account, which makes it more complex than the transformability. (Nyhuis et al, 2008). 1.2 Transformability Enablers In literature different enablers are discussed. (e.g. Hernández Morales, 2003; Koren & Ulsoy 2005; Heger, 2007; Nyhuis et al, 2008; Keijzer, 2008) The primary enablers that will be described in the following are compatibility, universality, mobility, scalability and modularization. (Nyhuis et al, 2008) The first enabler compatibility enables the creation of networks by defining interfaces. Another definition for compatibility is cross-linking ability. (Wiendahl et al, 2005) Compatibility in a broad sense means tolerance. The basic concept is applied for example in the field such as electrotechnology (e.g. Schwab, 1996), informatics (e.g. Stuckenholz, 2007), economics (Farrell & Saloner, 1987) as well as in psychology (Ickes, 1985). Besides the application in different research fields this is typical for the transformability enablers which are mainly discussed with a technical focus. The second transformability enabler is universality which contains the ability to adapt the dimension and design of a product or a technology to different requirements. (Nyhuis et al, 2008) The concept of universality is applied to production equipment, the related technological design and the changeover variety (Wiendahl et al, 2005) which can be optimized by the application of the modularization concept. (Abele & Wörn, 2004) Brehm defined this transformability enabler in contrast to adequacy as especially relevant in a highly unsecure environment where changes in the system-environment context cannot be predicted. (Brehm, 2003) Mobility means that a product, a plant or a process can be moved with small effort. In social science mobility means the movement of individuals between different classes or a change of profession and can be divided in vertical and horizontal mobility. (Kaelble, 1983) In the business context mobility is realized in the concepts of mobile work stations, square foot manufacturing (Wulfsberg et al, 2010) as well as the mobile plant (Schuh et al, 2004). Mobility in the context of human abilities means intelligent power and fast comprehension of new fields and areas. (Garavan & Coolahan, 1996) Scalability defines an expanding or reducing system characteristic, for example a flexible working time model. Scalability is an 4


applied concept in mathematics, production and especially in the information technology. In mathematics scalability means a simple form of coordination transformation. (Aumann & Spitzmüller, 1993) In the production research it is mainly applied in the flexible manufacturing system and reconfigurable manufacturing systems (ElMaraghy, 2006). Scalability for informatics architectures means that the architecture is scalable for one IT profile and in a planned capacity area, if there are practicable instantiation of the architecture (Brataas & Hughes, 2004). Modularity at least has some effects on the previously mentioned transformability enablers. It is a structuring method to reduce complexity in a system and therefore it is the base for the other enablers. Modularization is the restructuring of system, product, and service architecture into autonomous and manageable quantity of modules whose coordination is performed by non-hierarchical systems (Picot et al, 2003). As it was outlined previously the transformability enablers are strongly related to each other. Sudhoff formed a model of transformability enablers by defining the relation type either as encouraging or requiring or both of them (Sudhoff, 2007). In this paper the relation is represented by the arrows in figure 1, which means that one transformability enabler is an enabler for another. Mobility

Scalability

Modularity

Compatibility

Universality

Figure 1 : Model of transformability enablers (compiled by authors)

Mobility is generated by modularity, compatibility and universality but does not enhance transformability enablers itself. A compatible system can adjust to different environments as well as an universal system. Therefore a mobile system is universal and compatible, but a universal system is not necessarily mobile. Scalability is formed by compatibility, universality and modularity. Compatibility is necessary for an expandable and reducible system behavior as well as universality allows various system developments. Universality is enabled by modularity and compatibility. Compatibility supports the variety of processes. Modularity can be evaluated as a prerequisite for all other enablers. Mobility, scalability and 5


universality are advanced by the definition of modules that can be outsourced and broaden the action and reaction of a system. Compatibility evolves from defined interfaces that support a system to be more agile. For that reason the modularity was chosen to create an approach of transformability for SME and will be the focus of the next section.

2. CONFIGURATION POTENTIALS OF THE MODULARIZATION CONCEPT The following sections explain modularization in logistic system based on empirical research as well as conceptual work. The results of a modularization project are the basis for the following approach. The focus is on the implementation of logistics modularization in SME since modularization is one part of transformability. The approach outlines what steps have to be followed and how methods and tools have to be used. On the basis of a missing and distinct definition it is required to make clear what modularization of logistics means and what characteristics it consists of. In order to follow up that context the modularization basics and definition in a logistics system are given as follows. 2.1 Basics of Modularization The trade-off between flexibility and cost orientation is a classical situation in the production. On the one hand a standardized mass product can be produced for a reasonable price with low flexibility; on the other hand an individual product which can be easily adapted on customer needs causes high unit costs. Therefore a modularization concept provides flexible, reasonable and customized solutions (mass customization). The objective is to fulfill customer needs which can be satisfied by variants of services (variant configuration). The customer can combine different modules in order to individualize his product (Pine, 1993; Pine et al. 1993; Kotha 1995). The exchangeability of one module through another or the enlargement with a new module is the basic for flexible and quick reaction to customer need (Arnold et al, 2008). That fact is illustrated in the following figure 1. Based on a modular kit system the development effort for the customer can be reduced through shorter lead times, since a specific customer requirement can be fulfilled with existing standardized modules. The missing modules can accordingly be developed in order to meet the performance requirements. The illustration shows the trade-off between flexibility and cost-orientation and the classification of modularization.

6

RIRL 2010 - Bordeaux September 30th & October 1st, 2010 Flexibility vs. Cost Orientation

Standardized Product

Kosten Costs

niedrig low

Flexibilität Flexibility

niedrig low

Individual Product Modularization

hoch high

Costs

hoch high

Flexibility

Modular Company

M1

C U S T O M E R

M4 M6 M5

M2

M3

• Customized Offers (Variant Configuration) • Simplified module coordination by management • Flexibility through extension, decrease and exchange • Outsourcing potential

Figure 2: Flexibility vs. Cost orientation (compiled by authors)

2.2 Perspectives on logistics processes A complexity reduction is achieved by a modularization procedure while company areas are decomposed and visualized (Sattes et al. 2001). Defined resources and interfaces are assigned to modules and as a consequence the coordination of the modules can be managed easily (Burr, 2002). In addition, a new company structure is implemented which consists of modules reducing complexity by standardization. In the following section modularization is demarcated from well-known logistics perspectives in order to highlight the modularization potentials. Logistics as part of a functional organization describes logistics as a partial function within a department or as a separate one parallel to other departments in the hierarchy. However logistics within a functional organization leads to multiple interfaces within and outside the enterprise boundary which result to an additional coordination effort. Contrary to that process orientation in logistics means that company processes are cross-company and customerfocused orientated as for example the customer order process. Process orientation means reduced interfaces (Mayer, 2007). Compared to the described logistics perspectives (functional organization structure, process orientation) the modularization concept of logistic processes provides an alternative view. The organization structure of a company consisting of modules cannot be compared with logistics as part of a functional organizational structure. A modular structure seems to be very similar to the functional organization structure because the modules represent functional areas with interfaces which mean friction loss. In contrast to that two reasons are given, why modularization is not following the principles of a functional organization: 7


1. The determination of the modules is done by using module drivers and considering module-specific characteristics. The creation of modules is not coincidentally but it is a systematic approach to define standard modules. In addition, the modules can be formed by including processes from different functional areas. 2. The module interfaces are designed for friction loss free communication with other modules. That construction of interfaces is totally different from the functional organization structure. By the principle of process orientation, the logistical modularization is not far away. Process orientation is considering the logistics processes due to customer needs. In contrary to that the modularization is a summary of several sub-processes based on defined criteria. By combining different modules customer requirements can be fulfilled with individual services. 2.3 Modularization of a logistic system To answer the question according to the definition of the modularization of a logistic system, the following facts must be discussed: A logistic system consists of several elements which are related to each other. The modularization concept intends that some elements are less strongly linked up to each other while other elements are strongly interfaced. (Klinkner et al, 2005; Picot &Baumann, 2007) Elements with a strong relation to each other are grouped and called modules, because they could fulfill a process in a company as a closed entity. For example, the processes in the goods issue area (A) or the assembly of the product XY (B) can be modules A and B in the logistics. The assembly of the product XY is for itself a closed unit (module B) in which the single working steps stand in a stronger connection to each other than to the module A. The required information about the fulfillment of the customer order can be done by standardized interfaces as for example in the form of an information flow between both modules. The following figure outlines the modular system architecture as well as the characteristic of a module. A module is independent and delimited. It consists of own resources and defined interfaces to the environment and enables to be run in performance or cost responsibility (Mayer, 2007). A module is able to optimize itself and to achieve the agreed aims due to the independence as well as performance or cost responsibility. On the basis of the low dependence between the modules, modules are exchangeable, reusable and extendable (Burr, 2005; Matt, 2002;

8


Wildemann, 1998; Picot et al, 2003). With regard to the outsourcing of a module it can be considered as a black box with its defined interfaces (Aier & Schönherr, 2004). System Boundary System environment

Input

Output

Element Subsystem

Relation

Figure 3 : Modular système architecture (Klinkner et al., 2005)

The consideration of a module from SME point of view as a black box can foster innovation. With an interface specification SME permits Logistics Service providers (LSP) a free module design, as long as the interfaces are considered. This leads to a choice between two possible module operators. Either a specific module is operated by SME or the module is run by a service provider. The latter has the advantage that an increase of module performance can be achieved due to the specialization of the module operator (Burr, 2005).

3. TRANSFER OF MODULARIZATION CONCEPT TO THE LOGISTICS OF SME The previous section provides a detailed definition about the modularization in the context of logistics. The following section is structured in two different parts and it is based on empirical research. On the one hand a survey has been conducted in order to identify the relevant success factors for modularization projects as well as SME-specific requirements in terms of modularization guide are provided; on the other hand modularization approach is outlined with its methods, procedures and tools. Research’s objective related to modularization has been the development of a modularization guide supporting SME to create logistic modules and to specify the interfaces of each module. 3.1 Modularization success factors and requirements Success factors have been identified by conducting a literature analysis in order to figure out which factors are relevant for a successful modularization project. The identified success factors are differentiated by factors of organizational and functional manner, since there is 9


mutual relationship between organizational processes and functional knowledge in the context of medium to large projects. The organizational success factors can be grouped into three areas: management, employees and general methods. In contrast to that the functional success factors are categorized by logistics-specific success factors, structuring and the special methods due to the logistical issue. The validation of the identified success factors was carried out by a questionnaire and expert interviews. In the questionnaire the participants have been asked according to their personal assessment of the importance of individual success factors based on a scale of one - less important - until five – very important. The survey has been answered by six partners participated in modularization project who together represent a sample of ten. The survey was conducted in order to follow up the issue of modularization with the project partner as well as user recommendations and references for the implementation of a modular project can be derived from the result of the questionnaire. The survey results are summarized in the following table and show that the participating companies emphasize certain success factors. In particular, the functional success factors are related to the modularization and should be considered in the context of future implementation projects. The identified organizational and functional success factors are beneficial because these results can support the user in dealing with the modularization of logistic systems successfully. Success factors of modularization

Items

Organizational success factors Management

Realistic and clearly formulated definition of the project objectives

Competence of the project management (skills, experience, motivation)

Top-Management support

Employee Commitment in terms of project objectives and tasks

Qualification and experience

Communication and collaboration between internal project members and external partners

Organizational success factors – General methods

Assignment of responsibilities and competencies (project team)

Functional success factors – Specific Logistics Success Factors

Consideration of process orientation

Functional success factors – Structuring

Transparency through a structured, integrated approach

Functional success factors – Special Methods

Standardization: processes, interfaces, resources

Holistic consideration of the logistical system

Focusing on appropriate modules

Organizational success factors Employees

Table 1 : Identified success factors of modularization (compiled by authors)

The relevance of requirements for a modularization approach was also assessed in a questionnaire by the participating companies. The requirements of SME are structured into 10


the objective-related, application-related, monetary, personal and societal perspective. The most important objectives and requirements identified in the questionnaire are summarized in the following table. The results are based on responses of the same sample of SMEs. Sector

Manufactoring Industry

Objective-related perspective

Creation of value-added processes/services within modules

Better customer service due to increased quality of products/services

Increase competitiveness through high responsiveness

Realization of a transparent results

Easy applicability

Efficient progress of implementation project

Consideration customer-specific requirements

Achieve economies of scale

Evenly spread of costs within a cooperation

Increase the acceptance by the employees in terms of organizational change

Develop a practical and pragmatic problem-solving procedure

Concept is feasible with a few employees

Take into account of the social responsibility of management towards its employees

Application-related perspective

Monetary Perspective

Personal Perspective

Societal perspective

Table 2 : SME requirements for modularization approach (compiled by authors)

When a modularization procedure is developed and implemented the above requirements should be met in order to achieve an optimal result for the modularized, logistical system. For example, the respondents call for competitive advantage through a quick response capability if they implement modularization as well as they claim better customer service due to the increased quality in their processes. 3.2 Modularization approach In the course of this paper the approach of modularization, in particular the technique to logistics modules formation, is outlined and based on the findings from the literature review as well as taking into account the requirements of cooperating partners. The results are essential elements for the following conceptual work. For example, a combination of different methods has been selected which allows a simple, detailed and understandable approach which is flexible and holistic at the same time. Those are dependency matrices identified in the project such as Dependency Structure Matrix (DSM), the use of module drivers and approaches from the logistics modularization. The newly developed technique includes a statistical method for the module formation. The upstream and downstream processes of the module formation offer methods that have a high level of detail and ease of use. The modularization procedure is divided into two levels: processes and modules. It is an iterative process which implies process capture (1.), process visualization (2.) and process 11


evaluation (3.). After finishing process evaluation a two-step procedure follows within module formation (4.). The last two steps are the module design (5.) and module evaluation (6.) of the determined modules. The following figure illustrates on the one hand the modularization procedure and on the other hand methods, procedures and software tools for the respective process step. The methods behind each of the modularization steps are application-oriented which are partially supplemented by a software-based tool. The various methods of Modularization Procedure are briefly explained as followed. Modularization Procedure Process Level 1. Process capture

Process capture

2. Process visualization

3. Process evaluation

4. Module formation

BPMN‐Method

n.a.

Depedency Structure Matrix

Flyer

Feasibility analysis

Cluster analysis

Mindmap

KVP‐Procedure

Set up Matrices

Matrix sorting with Software

Module Designing

Checking Feasibility

Process capture model

Evaluation on basis of Module Drivers

Transfer into Excel

Module visualization

Constantly Quality Check

Gathering Documents

Consolidation of Matrices

Execution of Cluster analysis

X‐Mind

n.a.

Following Value stream Design

Process modeling

Module Design

6. Module evaluation

Process Evaluation

n.a.

Module Formation

5. Module Design

Process Visualization

Methods

Procedures

Module Level

Module Evaluation

Evaluation on basis of criteria (homogeneity)

Function Description

Determination of Cluster Grouping Software

n.a.

Sem Talk

Excel

PSM32, WinSTAT

Figure 4: Modularization Procedure (compiled by authors)

3.3 Process level The objective of the process capture is an actual analysis in a first step which is characterized by the capture of the total number of internal processes and activities in the form of process steps. This is a preparatory process for a visual illustration which precedes no activities as a precondition. If any process documentation exists, it can be considered for this step. The used procedures are based on the one hand on the value stream design, on the other hand a process capture model of relevant processes as well as a method for describing a process function. Gathered process documentation will be reviewed and verified in terms of relevance to the 12


current situation. Furthermore the activities which were not previously documented are captured. The processes and activities are documented in the form of a defined function description. Process visualization is pursuing the target of a visual representation which is based on the actual analysis. It is a precondition that the process steps are described in the proposed standard. The process visualization is based on the Business Process Modeling Notation (BPMN) which was developed as a model for business process modeling by the Object Management Group in 2004 (Krallmann & Hermann, 2007). The modeling of business processes is described by the four categories of elements: basic elements, connecting elements, responsibilities and context-dependent extensions. The data from the standardized process steps are transferred to the software tool (such as Sem Talk). The objective of step process evaluation is to create an overall matrix with processes that have been previously evaluated in terms of module drivers. This is done as a preparation in order to form the modules. As soon the visualization is finished an assessment of processes is carried out. For example, a spreadsheet like Excel can be used. After the processes have been added to Excel, an assessment is performed within matrices based on the module drivers. The result of step three is a total matrix with weighted overall processes. Evaluation on basis of Module Drivers

Set up Matrices • Four Matrices required • Set up n:n Matrix • Transfer Process steps

• Information Flow • Areal Layout • Resources

Consolidation • Addition Matrices • Consolidation into Total Matrix

Figure 5: Procedure for Process Evaluation (compiled by authors) Information Flow P1 P2 P3

P1 P2 P3 1

Areal Layout P1 P2 P3

Resources

P1 P2 P3 0

P1 P2 P3

Total Matrix

P1 P2 P3 1

P1 P2 P3

P1 P2 P3 2

0 = no dependency between two process steps 1 = dependency between two process steps

Figure 6 : Exemplary Evaluation (compiled by authors)

The composition of the total matrix is demonstrated as followed: In the matrix information flow booking goods receipt as one process step (P1) affects the process step putting parts in storage (P2) in regards to the information flow. A transfer order is generated in the warehouse 13


management system for storage, if the goods receipt has been posted. An information flow could be, for example, a status message in the system. In this context, a dependency between the two steps is at the level of information flow and the field will contain the value one. At the level of areal layout (matrix Areal Layout) there is no relation between the considered processes. Consequently, the field is either blank or zero. A comparison of the process steps P1 and P2 in the matrix Resources with regards to the module driver resources leads to the conclusion that there is a dependency. Specifically, at this point a scanner as well as an inventory management system are in use. Both equipments are technical resources. The dependency between both process steps results from the scanner that communicates with the warehouse management system. The warehouse management system generates the transfer order for parts storage. The last step is the total matrix. When the user adds all three matrices for this combination, you obtain the sum of two. The described procedure has to be done for all combinations of the matrix. 3.4 Module level The process step module formation at module level aims to form modules based on previous evaluated processes. As a method, a two-step procedure is used. For each step, an appropriate software tool is applied: PSM32 and WinSTAT. 1. Part: Sorting the total matrix to define first preliminary modules by the following steps

Sort algorithm

Determination of first modules

Preparation Cluster analysis

The modularization of logistics system pursues the target to group process steps with a high similarity within modules and to mark out from other processes with low similarity. The identification of an ideal amount of modules takes place in two steps starting with a DSM which is a dependency matrix and which was initially created as part of the process evaluation (total matrix). The objective is a matrix sort in order to determine the first temporary modules. The matrix is resorted along the main diagonal of the matrix using a sorting algorithm that orders the evaluated fields as a function of the valence (0, 1, 2, 3). Clusters are the sorting result that includes processes with close dependence. These clusters (modules) form the basis for the cluster analysis in order to determine the optimal amount of modules. The following example describes the procedure:

After sorting the Total matrix you could receive following results

14


Figure 7: An example - Sorted total matrix (compiled by authors)

Higher values are sorted across the diagonal

Highlight closed blocks (clusters)

In order to determine the optimal constellation of modules, cluster analysis is the second step

2. Part: Identifying an optimum number of modules according to defined similarity criteria (homogeneity criteria) by using a cluster analysis, a method of group formation. Although there are different procedures, the sequences of a cluster analysis is always similar. The sceme is illustrated as follows:

Determination of similarity

Merger algorithm

Determination amount of cluster (Backhaus et al., 2006)

For the cluster analysis the homogenity criteria shown in figure 7 were used. Quality

1 → no resp. minor Quality Requirement 2 → average Quality Requirement 3 → high Quality Requirement

Costs

1 → no resp. minor Cost Factor 2 → average Cost Factor 3 → high Cost Factor

Time/Dates

Human Resource

Decentralized Control

1 → no resp. minor time-critical meaning 2 → average time-critical meaning 3 → high time-critical meaning

1 → low qualification 2 → average qualification 3 → high qualification

1 → low requirement 2 → average requirement 3 → high requirement (PULL control)

Figure 8: Homogeneity criteria (compiled by authors)

The target in the next step module design is the content-related design and standardization of the formed modules. The collected and documented processes (step 1) and the visualized 15


processes (step 2) are gathered. Modules are identified and available (see Step 4). Mind map as a method can be applied for designing. Either flipchart or an optional software tool called XMind can be used as tools. The identified modules are designed as fliers supported by Mind map. If content-designed modules have been created (see step 5), step six Module evaluation consists of two methods: feasibility analysis and continuous improvement process (CIP or KVP). The identified modules are verified using set feasibility criteria. The goal is that the modules conform in terms of the economic efficiency, technical feasibility and compliance. The modularization is completed as soon as the feasibility analysis leads to a positive outcome. After a successful feasibility test the modules are regularly checked for their actuality and condition which is based on a certain procedure (CIP). If the feasibility test of the modules fails, the reason has to be determined. In that case two approaches can be proposed. The first approach would be a revision of the evaluated processes within the module and a re-run of module formation. Alternatively, if modules are considered to be outsourced the second solution could be the involvement of third parties such as a LSP. An advantage is the involvement and the support of the third party in an early process of implementation.

4. ADAPTION TO THE MODEL OF TRANSFORMABILITY ENABLERS An adaption of modularization supports SMEs to strengthen their performance und competitive advantage and to act more agile. The approach includes a modularization procedure and determination of success factors as well as SMEs requirements. Nevertheless modularization must be considered in the context of other transformability enablers. But how should transformability enablers be designed and deployed to raise the transformability of SME? Firstly for this objective the transformability enablers were explained and analyzed, after the analysis of the modularization approach a critical review of the interactions between modularization and the other transformability enablers as well as an implementation model to develop the logistics processes of SME. Therefore the main research question is answered by the following three questions: 1. What are the chances and obstacles for implementing transformability enablers? 2. How can an implementation of transformability drivers be accomplished? 3. How does the implementation of modularization influence the implementation of the other transformability drivers? 16


4.1 Chances and Obstacles of the approach The modularization approach was discussed with the participating SME and LSP. The experts confirmed that the structured approach supports all phases of the modularization project and enables a high level of process transparency and standardization. The holistic process analysis was the base for the discussion in the company and with the partners. Especially the detail level and the objectiveness were mentioned. All used instruments could be applied easily, but an obstacle was the broad range of different tools. It is stated that an integrated modularization tool which can be connected to existing systems would be necessary to cope with the main obstacle time. The time needed for implementing modularization in the whole company was evaluated as a problem for implementing and applying the modularization concept. Regarding their small resources the companies stated problems by implementing a theoretical approach in the whole company at once. Therefore discussions focused on how detailed analyses should be designed according to the company’s progress in the implementation of modularization. A phase model was developed. 4.2 Phase model of implementation The phase model shown in figure 9 refers to the stage of logistics development in the SME. It distinguishes between different stages such as functional organization, a process-oriented organization as well as the modularization levels 1 and 2. A functional structure represents the lowest level of modularization because of the multitude of interfaces between the different functions in an organization. A process-oriented structure is adjusted to the needs of the customer and the interfaces are optimized so that there is a continuous flow of information and goods. In phase 1 the company starts with the implementation of the modularization. In this process it is important that a best practice division is created, which means that the modularization approach is implemented in one selected division first. This can be a division or process that is planned to be outsourced or needs to be improved. In the division the application of the modularization concept can be tested in a limited complex environment and with manageable resource application. If the implementation is successfully finished the division can be a best practice model to learn, improve and then adopt the concept successively to other divisions. Completely modularized structure enables the system to adjust to customer needs with small effort.

17


Phase

Description

A0:

Logistics in a functional orientated organization

A1:

Logistics in a cross-functional und process-oriented organization

I:

Pilot project

Analysis and optimization of one division

Module specific design of the selected division

Outcome

II:

¾

One module

¾

Motivation: Outsourcing

¾

Area of action: short- to middle-term

Vision: Modularization of the whole enterprise by applying the modularization approach

After successful implementation and proving of the pilot project the approach will be implemented for further divisions

Outcome ¾

Structured modules for the whole enterprise

¾

Motivation: Outsourcing and variant configuration of the modules

¾

Area of action: middle- to long-term

Figure 9: Phase model (compiled by authors) 4.3 Effects and adaptability to transformability enablers Modularized logistics systems form the base for the realization and implementation of the other transformability enablers. Scalable logistics processes are enabled by the exact definition of the range of activities of one module. Those modules can be coordinated and advanced with partners to expand or reduce the own capacities if necessary. The outsourcing of modules can be relevant if there is a high demand that cannot be fulfilled with the own capacities or if there is a low demand which makes it necessary to involve a partner that can fulfill it more cost efficiently. The transformability driver universality refines the concept of modularization as well. The partners can specialize on the outsourced module and develop new solutions and applications. Based on the defined interfaces new approaches can be integrated with no additional effort. Therefore it is possible to offer a more universal service to the customer. Mobility, as the third transformability driver, is enabled by the integration of different partners in the value chain based on prepared exact definition of the module, the interfaces and the requirements. Although it has to be taken into account, that there exists a trade-off between universality and mobility. Universality is more efficient if there is a specific partner that operates the module in a long-term and has therefore the ability to develop innovative approaches. Contrary to universality, mobility is important if the module is price sensitive. For example, if it is price sensitive module, the innovative power of alternating partners will be lower than with one specialized partner. The fourth transformability enabler 18


is compatibility which is enhanced by the definition of interfaces. New solutions and processes have to be developed with standardized interfaces to the existing logistics modules. Therewith they can be integrated with minimum effort. It is shown that the concept of modularization has positive influence on the other transformability enablers. Nevertheless the creation of modules is based on a specific approach that makes it difficult to integrate other transformability enablers. On the other hand the general concept of defining a system and concrete approach based on the requirements and characteristics of the enablers can be adapted for the system of transformability enablers. This can be a research objective in future, because it is analyzed that a combined deployment of the transformability enablers has more potential than a single transformability enabler (Hernandez, 2003).

References Abele E., Wörn A. (2004), Chamäleon im Werkzeugmaschinenbau. Rekonfigurierbare Mehrtechnologiemaschine: ZWF 99, p. 152-156. Arnold D., Isermann H., Kuhn A. et al. (2008), Handbuch Logistik, Springer, Berlin. Aier S., Schönherr M. (2004), Flexibilisierung von Organisations- und IT-Architekturen durch EAI, in Aier S., Schönherr M. (Hrsg.), Enterprise Application Integration Flexibilisierung komplexer Unternehmensarchitekturen, GITO-Verlag, Berlin,. Aumann G., Spitzmüller K. (1993), Computerorientierte Geometrie, BI-Wissenschaftsverlag. Mannheim et al. Backhaus

K.,

Erichson

B.

et

al

(2006),

Multivariate

Analysemethoden:

Eine

anwendungsorientierte Einführung, 11. Aufl., Springer, Berlin. Belliger A., Krieger D. (2007), Wissensmanagement für KMU, Vdf hochschulverlag, Zürich. Bergmann L., Crespo I. (2009), Herausforderungen kleiner und mittlerer Unternehmen, in Dombrowski U., Herrmann C., Lacker T., Sonnentag S. (Hrsg.), Modernisierung kleiner und mittlerer Unternehmen – Ein ganzheitliches Konzept, Springer, Berlin Heidelberg New York, p. 5-29. Brataas G., Hughes P. (2004), Exploring architectural scalability, in Berichtsband des 4th international workshop on Software and performance, ACM Press, New 19


York, S. 125-129. Brehm C. R. (2003), Organisatorische Flexibilität der Unternehmung — Baustein eines erfolgreichen Wandels, Wiesbaden. Burr W. (2002), Service Enginerring bei technischen Dienstleistungen, eine ökonomische Analyse der Modularisierung, Univ.‐Verlag Wiesbaden, Gabler Verlag. Burr W. (2005), Chancen und Risiken der Modularisierung von Dienstleistungen aus betriebswirtschaftlicher Sicht, in Herrmann T., Kleinbeck U., Krcmar H. (Hrsg.), Konzepte für das Service Engineering: Modularisierung, Prozessgestaltung und Produktivitätsmanagement, Physica-Verlag, Heidelberg, p. 17-44. Cisek R., Habicht C, Neise P. (2002), Gestaltung wandlungsfähiger Produktionssysteme, ZWF, Jg. 97, H. 9, p. 441–445. Daschmann H.-A. (1994), Erfolgsfaktoren mittelständischer Unternehmen: ein Beitrag zur Erfolgsfaktorenforschung, Schäffer-Poeschel, Stuttgart. ElMaraghy H. A. (2006), Flexible and reconfigurable manufacturing systems paradigms, in International Journal of Flexible Manufacturing Systems, Vol. 17, p. 261-276. Ernst D. (1999), Internationalisierung kleiner und mittlerer Unternehmen: Kooperationsformen und Außenwirtschaftsförderung, Dt. Univ.-Verl., Gabler Verlag, Wiesbaden. Farrell J., Saloner G. (1987), Competition, Compatibility and Standards: The economics of horses, penguins and lemmings, UC Berkely, Departement of Economics. Garavan T.N., Coolahan M. (1996), Career mobility in organizations: implications for career development, Journal of European Industrial Training, Vol. 20. p. 30-40. Goldmann S. L. (1996), Agil im Wettbewerb, Springer Verlag, Heidelberg. Hernández M. R. (2003), Systematik der Wandlungsfähigkeit in der Fabrikplanung, Dissertation, University Hannover. Heger C. L. (2007), Bewertung der Wandlungsfähigkeit von Fabrikobjekten, Dissertation, University Hannover.

20


Hildebrand T. (2005), Theoretische Grundlagen der bausteinbasierten, technischen Gestaltung wandlungsfähiger Fabrikstrukturen nach dem PLUG+PRODUCE Prinzip, Dissertation, Technische Universität Chemnitz. Ickes W. (1985), Compatible and incompatible relationships, Springer, New York. Kaelble H. (1983), Soziale Mobilität und Chancengleichheit im 19. und 20. Jahrhundert, Göttingen. Keijzer W. C. (2008), Wandlungsfähigkeit von Entwicklungsnetzwerken: Ein Modell am Beispiel der Automobilindustrie, Dissertation, Technische Universität München. Koren Y., Ulsoy G.A. (2003), Reconfigurable Manufacturing Systems, in 2nd International CIRP Conference on Reconfigurable Manufacturing, University of Michigan, College of Engineering. Klinkner R., Mayer A.et al.(2005), Modulare Logistik ‐ Ein Lösungskonzept zum Management von Komplexität in dynamischen Netzwerken, in Industrie Management, Vol. 21, Nr 5/2005, p. 33–36. Kotha S. (1995) Mass customization: implementing the emerging paradigm for competitive advantage, Strategic Management Journal 16, pp. 21-42. Krallmann H. (2007), Systemanalyse im Unternehmen. Prozessorientierte Methoden der Wirtschaftsinformatik, 5. Edition, Oldenbourg, München. Mayer A. (2007), Modularisierung der Logistik. Ein Gestaltungsmodell zum Management von Komplexität in der industriellen Logistik, Disseration, Technische Universität Berlin. Matt D. (2002), Planung autonomer, wandlungsfähiger Produktionsmodule, in ZWF Zeitschrift für wirtschaftlichen Fabrikbetrieb, Jg. 97. Möller N. (2008), Bestimmung der Wirtschaftlichkeit wandlungsfähiger Produktionssysteme, Dissertation, Technische Universität München. Morgan C. (2004), Structure, speed and salience – performance measurement in the supply chain, in Business Process Management Journal, 10/5, p. 522-536. Noffen D., Klußmann J.H., Löllmann F. (2005), Komponenten und Aufbau einer wandlungsfähigen Fabrik, in Wiendahl H.-P., Nofen D. et. al. (Hrsg.), Planung modularer Fabriken, Carl Hanser Verlag, München.

21


Nyhuis P., Reinhart G., Abele E. (Hg.) (2008), Wandlungsfähige Produktionssysteme. Heute die Industrie von morgen gestalten, Garbsen, PZH Produktionstechnisches Zentrum. Picot A., Baumann O. (2007), Modularität in der verteilten Entwicklung komplexer Systeme: Chancen, Grenzen, Implikationen, in Journal für Betriebswirtschaft 57 Picot A., Reichwald R., Wigand R. T. (2003), Die grenzenlose Unternehmung – Information, Organisation und Management, 5. Edition, Gabler Verlag, Wiesbaden. Pimmler T.U., Eppinger S.D. (1994), Integration Analysis of Product Decompositions, Proceedings of the ASME Design Theory and Methodology Conference, Minneapolis. Pine J. (1993), Mass Customization : The New Frontier in Business Competition, Harvard Business School Press. Pine J., Victor B. and Boynton A. C. (1993), Making mass customization work, Harvard Business Review. Ramesh G., Devadasan S. R. (2007), Literature review on the agile manufacturing criteria. Journal of Manufacturing Technology Management 18. p. 182-201. Sattes I., Brodbeck H., Bichsel A., Spinas P. (2001), Praxis in kleinen und mittleren Unternehmen: Checklisten für die Führung und Organisation in KMU, vdf Hochschulverlag, Zürich. Schuh G., Merchiers A., Kampker A. (2004), Geschäftskonzepte für global verteilte Produktion, Ergebnisse des Verbundforschungsprojektes „MobiFak – Geschäftskonzepte für mobile Fabriken in wt Werkstattstechnik Online, Jg. 94, Heft 3, p. 52-57. Sudhoff W. (2007), Methodik zur Bewertung standortübergreifender Mobilität in der Produktion, Dissertation, Technische Universität München. Schwab A. J. (1996), Elektromagnetische Verträglichkeit, Springer, Berlin. Shewchuk J. P., Moodle C.L. (1998), Definition and classification of manufacturing flexibility and measures, in International Journal of flexible manufacturing systems, Nr.2, p.289-328. Straube F. (2007), Die Bedeutung der Logistik in Wissenschaft und Wirtschaft, in Hausladen I. (Hrsg.), Management am Puls der Zeit – Strategien, Konzepte und Methoden, Band 2, Produktion und Logistik, TCW Transfer-Centrum München, p. 997-1014.

22


Stuckenholz A. (2007), Kompatibilitätsbewahrende Komponentenupdates, Disseration, Fernuniversität Hagen. Tichy N. (1983), The essentials of strategic change management, Journal of business strategy Westkämper E., Zahn E. (Hg.) (2009), Wandlungsfähige Produktionsunternehmen. Das Stuttgarter Unternehmensmodell, Springer, Berlin. Wiendahl H.-P., Nofen D., Klußmann J. H., Breitenbach F. (2005), Planung modularer Fabriken, Hanser, München. Wildemann H. (1998), Die modulare Fabrik – Kundennahe Produktion durch Fertigungssegmentierung, 5. Edition, TCW-Transfer-Centrum GmbH, München. Wulfsberg J. P., Redlich T., Kohrs P. (2010), Square Foot Manufacturing: a new production concept for micro manufacturing, Production Engineering, Volume 4

23



Fulfillment systems in Multi-Channel Retailing - Customer Expectations and Economic Performance

Gerald Lang BEM Bordeaux Management School, France Ecole Polytechnique PREG-CRG, France [email protected]

Abstract: Increasingly, store-based retailers are opening an additional online sales channel and becoming multi-channel retailers. The integration of these different channels raises the question how to redefine the strategic marketing elements and the operations, as the two channels have different constraints and require different competences. This multi-channel retailing has major impacts on the operations and the supply-chain management. Order fulfillment for the customers using the different sales channels is a key challenge, as customers are directly impacted from the performance of this process. In this paper, we will highlight the different strategic elements and operational challenges to be addressed by a multi-channel retailer. In order to evaluate the match between a fulfillment system and a multi-channel business model, we propose a framework analyzing the overall performance of different fulfillment systems regarding the two dimensions customer expectations and economic performance. Key Words: Multi-Channel, Retail, Fulfillment, Distribution

1


INTRODUCTION A growing number of traditional store-based retailers are becoming multi-channel retailers by opening an additional online-channel to offer their customers multiple ways of buying and interaction. At the same time, former pure online-retailers are opening stores or physical pickup points to allow their customers physical interaction and product pick-up. Multi-channel retailing got primary attention principally in the field of multi-channel marketing (e.g. Wirtz, 2008). Marketing literature and scholarly research address multichannel marketing mainly focusing on customer behavior on the one side (e.g. Venkatesan et al, 2007) and the conflicts between channels on the other side (e.g. Falk et al, 2007; Avery et al, 2009). Several authors (e.g. Schröder, 2005; Schobesberger, 2007; Heinemann 2008, 2009) analyze different multi-channel strategies and implications on a retailers' marketing mix. Whereas operational implications of online retailing have received attention since the end of the 1990s (e.g. Turban et al, 2009; Swaminathan & Tayur, 2003), specific analysis of operational implications of multi-channel retailing remains scarce. Agatz et al (2008) reviewed the specific supply chain management issues of internet fulfillment in a multichannel environment. They identified an important number of topics relevant for efulfillment, but recognized that there are only very few dedicated models on multi-channel retailing. Multi-channel retailers with different business models are using different fulfillment systems to deliver the products to their customers: The online shop of the French retailer Fnac (cultural and electronics products), fnac.com, started its online activity completely separated from its store activity, with different warehouse locations and fulfillment organizations. They gradually integrated their logistics activities and are operating today a fully integrated warehouse, preparing deliveries for the stores as well as for the individual online customers. Tesco.com, the online shop of the major UK retailer Tesco, prepares the customer's orders exclusively in the Tesco store next to the customer (Smith & Sparks, 2009). So does monoprix.fr, the online shop of the French city-center supermarket chain Monoprix. These observations suggest that a specific multi-channel business model may induce a specific, most suitable fulfillment system. If this is the case, it should then be possible to evaluate how a fulfillment system fits a given business model. 2


From an operational point of view, fulfillment systems have to be mainly time- and costefficient. From a strategic marketing point of view, the fulfillment system has to be able to meet the customers' expectations created by the chosen business model. We therefore propose to develop a qualitative framework to evaluate how a specific multichannel fulfillment system matches the business model of multi-channel retailers both in terms of the expectations of their customers on the one hand and the economic performances on the other hand. This framework should serve as basis for further empirical studies. Our paper is organized in two parts: In the first part, we will highlight the main characteristics of multi-channel retailing and present the major elements of the retailers' strategy, the retail mix, which have to be adapted to a multi-channel retailing strategy. 1 We then will explore the specific operational challenges to be faced when operating a multichannel activity. This first part will be completed with the presentation of the five major multi-channel fulfillment systems used to deliver products to the customers. In the second part, we will propose a qualitative analysis framework in order to evaluate the overall performance of a given fulfillment system, combining the two dimensions 'customer expectations' and 'economic performance'. The parameters of the framework as well as its applications will be based on the analysis of the strategic and operational challenges developed in the first part. Applying the analysis framework to the five major fulfillment systems leads us to discuss how each of the systems impacts the different parameters. We conclude our paper presenting managerial implications and suggested future research avenues to provide evidence if our framework is valid for our purpose.

Methodology Our analysis is based on the one hand on our longitudinal industry observation of retailers' practices, completed with the analysis of company information and press articles. Moreover,

1

In our paper, we will concentrate on multi-channel retailers operating at least several physical stores and an online sales channel under the same brand and with the same types of products. Moreover, we focus on the distribution of physical products, as dematerialized products do not need a physical distribution. We exclude as well specific configurations (which may nonetheless be dominant in particular sectors), like delivery from manufacturers or 'complex assortments' (de Koster, 2002) with bulky items and temperature sensitive goods.

3


we conducted a series of interviews with French and UK senior operations managers and consultants in the retail sector. This observation allowed us to identify five major multichannel fulfillment systems. To be able to evaluate these fulfillment systems regarding the two dimensions 'economic performance' and 'customer expectations', we have to identify the relevant parameters to integrate into our proposed framework. To do so, we analyzed Marketing and Operations general literature and scholarly research. Exploration of the main strategy elements and operational challenges suggests number of parameters, indicating economic performance and customer expectations. We completed the identified parameters for evaluating customer expectations with dimensions from the ServQual and NetQual scales, measuring service quality perception from the customers. These parameters of our framework will be the dependent parameters, whereas the analyzed fulfillment systems will be the independent parameters. The exploration of the strategy factors and economic performances in the first part will allow us to apply the framework to the different fulfillment models by assessing the chosen performance parameters for each of the fulfillment systems. This shall provide a global evaluation for the each of the analyzed fulfillment systems.

STRATEGY FACTORS IN MULTI-CHANNEL RETAILING The objective of this chapter is to summarize the context of multi-channel retailing as well as the main drivers for customer expectations and economic performances, which will enable us to identify the relevant parameters to use in our proposed framework. We will first present the main characteristics of multi-channel retailing. We then will investigate how the business model of a multi-channel retailer, characterized by different elements of the retail mix, will have to be adapted to a multi-channel strategy. An exploration of the major operational challenges for a multi-channel retailer will conclude this chapter.

Multi-Channel Retailing: Characteristics & Customer Behavior Worldwide retailing is still dominated by the traditional store retailing, as 95% of food retailing and 80% of non-food retailing is realized through physical store-based retailers (Dioux & Dupuis, 2009), and online channels still represent only a small part of the total sales of the retailers. Nevertheless, the motivation to open online sales-channels is strong due to the fact that e-commerce is realizing important annual turnover growth rates of about 30% for the 4


last few years, whilst most other retail forms stay stable or even decline. It is therefore interesting for a store retailer to invest in the creation of an additional online sales channel. To implement a multi-channel strategy, retailers have to decide how to adapt their business model by adjusting the elements of the retail mix. At the same time, they have to face and resolve numerous operational challenges. In multi-channel retailing, customers expect at least the same level of service, counsel, and information on every channel (cf. Heinemann, 2008). They expect to have identical customer loyalty programs across all channels, to be able to pay and pick up in the shop a product ordered on the web site, or to be able to return and exchange a product bought on the web in a store (Berman & Thelen, 2004). This channel hopping, which is a legitimate desire from a customer's point of view, may nevertheless prove to be very difficult to implement for the retailer (Vanheems, 2009). At the same time, online shoppers transfer their perception and experiences from a physical store to the website of the same brand and vice versa. Customers tend to transpose a bad experience in one channel to the other channels of the same distributor (Heinemann, 2008). As soon as their expectations are not fulfilled, they may adapt their purchasing habits, in the worst case stop buying with this retailer at all. It is therefore vital for a multi-channel retailer to make sure that every channel always meets the same level of expectations of the customers.

Multi-Channel Retail Mix The Retail Mix consists of the mix of variables, including geographical location, merchandising, communication, price, services, product range and the personnel, which form the overall strategic marketing components of a retailing activity. It is the right mix of all these factors which determines if a retail concept is viable and successful. Compared to a store retailer, the online sales channel is not only a new channel, but rather a new business model, where experiences from the traditional (brick and mortar) store channel may not be valid any more (Heinemann, 2008). Decisive success factors in a store like store location, sales personnel and material management in the shop are not the same success factors in an online sales channel. Retailers which decide to open an e-commerce activity beside their existing store sales channel have to make sure that the co-existence of the two sales channels is working well. They have to decide how to adapt the different elements of the retail mix to each channel, 5


which product strategy to chose, which prices to apply, how to maintain a consistent image across the channels or what role to assign to each channel, to name a few (cf. Berman & Evans, 2009). We will briefly describe the major issues of some of the most important variables of the Retail Mix, assessing their relevance and importance in the store, the online channel and the multichannel strategy. Product Mix The product mix is the major variable determining the activity of a retailer. It establishes if a retailer is specialized in selected product categories or a generalist. If a store based retailer opens an additional online channel, he has to decide which product mix he offers on this new channel. He can offer the identical product mix, a combination or part of his existing product mix or a completely different additional product mix on the new channel. Location of the stores A physical store's success is to a great extend influenced by its location. In order to determine if the location can attract the store's target customers, a store based retailer has to analyze socioeconomic data, transport access, mix of competitors, costs and regulation issues. These parameters, decisive for a physical store, have generally no influence on a pure online sales channel, as access to the websites is independent from a geographical location. In a multi-channel strategy, however, the store location remains important; the density of the store network can be an important factor for offering store pick-up, returns and after sales services in the stores for their online customers. Pricing The pricing policy applied in a store may take into consideration the environment of the store (socioeconomic characteristics of the geographical catchment area) and the price levels of the local competitors. Such a local pricing policy which can be applied in a store is incompatible with the global price transparency in an online channel and the exposure to price comparisons. The pricing policy of a multi-channel retailer encounters specific difficulties: the global price transparency on the internet, the specific cost structures of the store and the online channel,

6


the existence of specific offers and the different means of payment render it difficult to apply a consistent pricing policy across all channels. Especially the price transparency makes it difficult for a retailer to offer different prices for the same product on different sales channels. Service, store pick-up and return A multi-channel service offer can be composed of special delivery options (delivery speed, delivery timing, information on order status), the exchange policy, return policy, refunding, after sales service, payment options and consumer credits. Multi-Channel retailers can offer either to deliver the order to the customer's home or to let the customer pick-up the order in one of the stores. This service allows the customer to pick up the order whenever he likes, even sooner than with home delivery.2 Allowing customers in-store return or exchange can avoid the cost and time of returning via mail. At the same time, this offers an important possibility for the retailer to create a personal contact between the client and the sales personnel to increase customer loyalty and to generate additional sales. Personnel The sales personnel play a decisive role in stores. They welcome clients, provide counsel and advice and can influence sales. Their behavior and attitude contribute in an important way to the atmosphere and ambiance of the store and its concept. This important parameter in a store is irrelevant in a pure online channel, except for telephone hotline contacts or the delivery service personnel. In a multi-channel strategy, however, the store personnel have to be able to orient and counsel the client using different channels. It is very important for the client not to perceive any differences or barriers between the channels.

Operational Challenges of Multi-Channel Retailing In general, the operational organization of a retailer is mainly focused on efficient store replenishment, lowering costs and shortening the product turnover time.

2

In case of unavailable items, the retailer can chose to send the missing items directly to the customer and only reserve the available items in the store for pick-up. In this case, the customer experience will be negative, as the customer's decision to pick up his order in the store can't be fulfilled.

7


To offer an online channel to its customers, a retailer has to be able to manage its operations in a very different way from its physical store activity. E-fulfillment may be considered as one of the most expensive and critical operations of Internet retailers (de Koster, 2002), logistics and stock management becoming core competencies (Benghozi, 2001). Keeping stock of a high number of different references, making it available, picking and packing, sending and delivering quickly at a low cost are the numerous constraints to respect. In the following, we will present and discuss the main operational issues, which are decisive for the overall performance of the multi-channel retailing. Product Availability Product availability is one of the most important parameters for customer satisfaction (Browne, 2009). A store customer usually can compare the products present in the store. Availability is indicated through the simple presence of a product in the store shelve. A customer can ask the sales personnel if there is any stock of the product or when it will be available again. An online client needs to have all this information indicated on the website together with the product information. The online channel offers the possibility to display the remaining stock of a chosen product, or even, in a multi-channel context, the remaining stock and the location of the store where the product is available for testing, purchasing or withdrawal. Displaying the availability of products or the delay of their replacement is an important marketing tool impacting the sales and influencing customer satisfaction. Stock keeping Stores usually receive deliveries either directly from the manufacturer or from one or several logistics platforms with either warehousing or cross-docking function. As online orders may consist of different products, composing the orders from different warehouse locations may prove to be difficult and inefficient in terms of cost and delivery time. A central warehouse and fulfillment center may therefore, at a first view, be more adapted to the online channel. However, multi-channel retailers may as well pick the items for an online order directly in the stores. For an online order, the retailer has some flexibility where from he delivers the products to his customers, as he can use multiple stock locations and therefore offer a wider product range than the one contained in a single warehouse, as multiple channels may share their 8


inventories. In-store inventories can therefore be available for online customers' orders (Agatz et al, 2008). Centralizing stock allows easier monitoring of the stock levels for different products. If a retailer is serving its online clients (totally or partially) from stock in the stores, the overall stock level on these different locations is difficult to monitor. This is the same problem for multi-channel retailers who offer the possibility to their customers to order online and withdraw the product in a nearby store. As information on product availability is a crucial parameter for the customers buying decision, the accuracy of the stock level information is very important. Order taking and fulfillment In an online channel, the customer enters the type and number of products he wants to acquire and enters the payment information (usually done by credit cards), the online channel therefore becomes a self-service channel (Falk et al, 2007). All this information entered by the customer triggers the order preparation and fulfillment process. Retailers once delivering their stores from centralized distribution centers (often with complete boxes or pallets) have to re-think the way orders are prepared, as their organizations are not designed for small individual customers' orders. They must therefore be reorganized to treat these orders (de Koster, 2002), implementing a process of single item picking and packing. As mentioned above, another option for the retailer is to pick the items in the stores, where the products are available on the shelves, and to prepare the order in the shops. Furthermore, orders may be consolidated from multiple stock locations (for example from a warehouse and a store), which needs a specific process. The more complex the fulfilment process is organized, the more difficult it becomes to respect the quality and efficiency of this process. Delivery to the customer Besides the availability of the product, the different delivery options for a product are an important parameter for the online channel client. Different delivery service levels offered like delivery in 24 hours, delivery to a pick-up point, national post service etc. are becoming a product differentiation factor of the retailer.

9


Delivery to the customers' homes, which is a distinctive element of an online channel, implies to deal with cost and organizational issues. The usually small transactions have to be managed and delivered in an efficient way, while respecting the promised delivery timing. The costs of the home delivery can be totally or partially paid by the customer. This becomes an important parameter of choice for the client, as no direct delivery costs appear in the store channel. On the other hand, the costs of delivery can be used as a marketing tool for special offers ("your next order free of charge") or minimum order amounts ("delivery free of charge for every order above 100 Euros"). Delivery becomes therefore a service parameter, a price parameter and a promotional parameter. Reverse logistic Store customers can usually see, touch and try the goods before purchasing. Returns are limited and very often linked to an exchange of the product, where the returned product is put back into the shelves. Defect products are reimbursed. As online customers cannot see, touch and try the products, online retailers usually offer good conditions for the customer to return the products they don't like (due to legal and regulatory obligations, but as well as part of their customer service offer). Return handling must be quick in the eyes of the customers, either to get a product exchange or to get reimbursed and the customer account re-credited. The process must as well be efficient as it is a very manual process and impacts the stock management of the returned products (Min et al, 2006). A physical store channel can be an easy way in the customer's eyes to return the products.

MULTI-CHANNEL FULFILLMENT SYSTEMS To operate the multi-channel strategy, retailers have to set up a fulfillment system to deliver products to the customers. This fulfillment system, organizing the process from order intake to delivery to the customer, has to match the business model and to reply to the operational challenges. Industry players are using different fulfillment systems, and several authors suggest typologies of fulfillment schemes, focusing on the locus of order preparation (de Koster, 2002; Durand, 2008). Multi-channel retailing per se intends to offer the clients the choice where and how to order and where and how to get the products delivered. 10


The major choice for the client is to shop in the store or shop online. If shopping online, the client can furthermore opt for delivery at home or pick up his products in one of the retailer's stores. The retailer has to set up a fulfillment system able to respect the client's choice of the delivery mode, either by creating a new logistics infrastructure, or using the existing one, by adapting the existing processes. In order to be able to analyze the fulfillment systems from a customer expectation's perspective, we decided to distinguish the fulfillment systems by the delivery option chosen by the customer: Clients may either choose the direct home delivery for their purchases completed on the online channel or choose to pick up their purchased items in a physical store of the retailer. This in-store pick-up is a delivery option only multi-channel retailers can offer and which may be chosen to 30-40% of the customers at some multi-channel retailers (Berman & Thelen, 2004). A client going into a store in order to buy a product, searches, compares, picks the product out of the shelve and pays the product. Single-channel store shoppers may therefore not be affected by the multi-channel fulfillment organization. 1. Direct Home Delivery Direct home delivery can be completed either by the retailer's own delivery organization, a delivery logistics provider or via an external pick-up point network3. Before entering in this 'delivery pipeline', products have to be picked, prepared, packed and expedited, which can be done either in a central warehouse or distribution center, in a store, or in multiple locations (central warehouse and one or more stores).

3

We consider a delivery via an external pick-up-point network (usually neighborhood shops linked to a network which offers the retreat of parcels for the clients of different retailers) like a home delivery, as the fulfillment process is quite identical, with a delivery of (sometimes pre-sorted) parcels to the pick-up point network's injection hub. Although this is influencing the ease of access, this way of delivery cannot provide a contact with the retailer as in his own store.

11


1.1 Direct Delivery from central warehouse

Picking, preparation and packing of ordered items can be done from a central warehouse stocking the products (Figure 1.1). This central warehouse can either be a purpose-built entity to serve exclusively the online channel or the existing warehouse for the store operations (in which case the processes in the warehouse have to be adapted). Figure 1.1: Direct delivery from central warehouse

1.2 Direct Delivery from a store

Picking, preparation and packing of the ordered items can be done inside a physical store out of the shelves. In this case, operators pick the products into a specific order preparation cart from the shelves and then pack and send the completed order to the customer (Figure 1.2).

Figure 1.2: Direct delivery from a store

1.3 Direct Delivery after multi-stock consolidation

If the desired articles of an order are not available in one place, picking of the ordered items has to be done in a central warehouse as well as inside a physical store from the shelves. The picked items have then to be consolidated, prepared and packed in one site (Figure 1.3). In some cases, the multichannel

Figure 1.3: Direct delivery after multi-stock consolidation

12


retailer may opt for sending the order in two different parcels, to avoid the costs and delay of the consolidation. 2 Client Pick-up in the store A customer ordering online may choose to pick up his products in one of the retailer's stores. In this case, products have to be picked, prepared, packed and kept ready for pick-up in the store. 2.1 Direct in-store order preparation

If all the products of the customer's order are available in the store chosen by the customer to pick-up, the picking, preparation and packing of the order can be done immediately in the store (Figure 2.1).

Figure 2.1: Direct in-store order-preparation

2.2 In-store order consolidation from multiple stocks

If some of the products of the customer's order are not available in the store chosen by the customer for pick-up, the unavailable items must be ordered from the central warehouse (or other stores) to complete the order. The preparation and packing of the order can be done only after consolidation of all items in the store (Figure 2.2).

Figure 2.2: In-store order consolidation from multiple stocks

(A specific case of this in-store order consolidation can be a complete order preparation in a central warehouse sent to the store only to be picked up by the customer, without further adding any product in the store). 13


THE ANALYSIS FRAMEWORK In this part, we will first determine the parameters to integrate in our proposed framework. We then will apply this framework to the five abovementioned fulfillment systems and discuss the results. We will conclude this part discussing the limitations of the framework, suggestions for further research and managerial implications. Customer expectations and economic performance Having analyzed the main strategy factors and operational challenges, and identified the five major fulfillment systems, we now have to determine the parameters to include into our proposed framework in order to evaluate these fulfillment systems. The different fulfillment systems lead to different levels of complexity, of costs and of product availability and therefore different levels of customer satisfaction and economic performance. To explore how a fulfillment system fits a given business model, we have to evaluate the overall performance of the different multi-channel fulfillment systems, combining the two angles of analysis: On the one hand, we analyze to which extend a given fulfillment system is able to meet the customer expectations. On the other hand, we analyze the economic performance of each model. In our proposed framework, we will confront the different fulfillment systems to several parameters which are relevant for the customer expectations and the economic performance. The fulfillment systems may be considered as the independent variables of our analysis, whereas the different parameters of our framework represent the dependent variables. The identification of these parameters results from the exploration of the main strategy elements and operational challenges. Concerning the economic performances of fulfillment, we propose to assess different dimensions of the overall fulfillment efficiency, combining widely used indicators to describe logistics and fulfillment performances (e.g. in Agatz et al, 2008): ‐

Stock efficiency (indicating reliable sales forecasts and optimized product

replenishment to reduce stock-risk, capability of monitoring stock levels across different stock locations and the capability of balancing stock between different locations) ‐

Picking efficiency (indicating efficient organization of stock and picking zones for

picking and order preparation) 14


‐

Delivery cost efficiency (indicating the total cost of delivery from the stock(s) to the

customers' home or to the desired store for pick-up) ‐

Channel organization efficiency (indicating the information exchange between sales

channels and real time availability of data) ‐

Fulfillment infrastructure costs (indicating the necessary investments, rents or

outsourcing costs for the fulfillment infrastructure) ‐

Return handling efficiency (indicating the speed and quality of the return handling

process, from product reception to exchange or reimbursement for the client)

Analysis in the first part of this paper suggests three parameters indicating the customer expectations: ‐

Product availability (indicating the availability of the desired products)

‐

Speed of delivery (indicating the time between the order placement and the reception

of the products by the customers) ‐

Ease of access and return of products (indicating the ease for a client either to get the

products delivered/to pick them up or to return/exchange the products received) In order to assess more precisely how a fulfillment system corresponds to the customer expectations, we completed these parameters with parameters from the ServQual framework, developed by Parasuraman et al (1988) to measure service quality. Nevertheless, this traditional service quality may not be applicable in an online environment (Parasuraman et al, 2005). Bressolles (2006) developed the NetQual framework to specifically assess electronic service quality. As multi-channel retailing provides both 'traditional' and 'online' service experiences, we have to combine these two approaches in order to identify all the relevant dimensions to customer expectations and retain only those which may be directly impacted by the fulfilment model. Therefore, we selected the following dimensions out of the ServQual and NetQual scales for our analysis: ‐

Quality and quantity of information (NetQual) (indicating the availability and

correctness of the information provided in terms of product availability, stock level, delivery times etc. either on the website or in the stores)

15


‐

Reliability and respect of commitments (ServQual & NetQual) (indicating the ability

to deliver the products in terms of quantity, price, description and timing) ‐

Offer (NetQual) (indicating the lengths and the width of the product range available on

the different channels) ‐

Assurance (ServQual) (indicating the competence and friendliness of the personnel

usually in the store for buying, pick-up and exchange transactions of the client).

We did not retain those dimensions having only an indirect link with the fulfillment systems. These elements constitute the analysis framework we propose to analyze the overall performance of specific fulfillment systems.

Application of the Analysis Framework We are now able to apply this proposed framework to the five identified fulfillment systems according to our objective to evaluate the fit between a multi-channel retailer's business model and a specific fulfillment system. To be able to analyze each of the abovementioned fulfillment systems, we will explore the sensitivity of each of the parameters regarding the main characteristics of the different fulfillment systems.4 We apply numerical scores to each parameter, depending if the efficiency of this parameter is high (2), medium (1) or low (0). This qualitative evaluation, based on the discussion of the different elements of the retail mix and the operational challenges, aims to compare the relative efficiency of one parameter in different fulfillment systems. We do not intend to provide an absolute performance test. Quality and Quantity of Information Customers expect correct information, especially on pricing and delivery costs of the different channels, but as well on other parameters like product availability, speed of delivery and product characteristics. The retailer's challenge is always to have the right information available on every channel.

4

Each of the above identified parameters expressing customer expectations and economic performance of a specific fulfillment system has multiple implications on other parameters. It is to retain that these parameters may be interdependent (e.g. the width of the product offer impacts the product availability, the picking efficiency and the fulfillment infrastructure costs).

16


Using a centralized stock provides the highest chance to have the right information available (score: 2), whereas an order preparation from stores induces the problem of obtaining the right information on the store stock level (score: 1). If the order preparation has to be completed on several stock locations (central warehouse and stores), it is very difficult to obtain correct information and the quality of information may be low (score: 0). Reliability and Respect of commitments Customers expect retailers to keep their promises, especially in terms of delivery time and delivery service. Central warehouse order preparation should facilitate this, as the quality of information is high and the fulfillment process can be optimized (score: 2). Order preparation in the stores suffers from difficulties in keeping a real-time stock level of all the products. Therefore, store preparation may guarantee only a medium level of reliability and respect of commitments (score: 1), whereas multi-stock order preparation may only provide a low level (score: 0). Offer Customers expect to find the offer they expect on every channel, a large or narrow product range with the appropriate depth of assortment. The retailer has to make sure to be able to offer the products customers expect on every channel. Order fulfillment at a central warehouse or from multiple stocks should be capable to provide the entire range of the products (both score: 2), order preparation at the store is limited to the only product assortment in stock at the store (score: 0). Product availability The customer expects that the products he intends to buy are available in the channel he desires to use. A central warehouse order preparation may provide only a medium level of product availability (score: 1) compared to the other options, as the availability depends on the stock keeping and forecasting precision of this single warehouse. Stores can only guarantee a low level of product availability because of the small number of stock for every referenced product and the continuous buying from customers in the store (score: 0). Multi-stock order preparation may use the stock of different entities (stores and central warehouse) and therefore guarantee a very high level of product availability (score: 2), especially in case of a high number of articles per customer order. 17


Speed of delivery Customers usually expect a rapid delivery of the products they bought and to be informed about the delivery timing. Retailers therefore have to implement a fast and reliable delivery process, which may imply an external delivery service provider. A central warehouse usually needs to bridge the distance to the final customer, which may be long and therefore time-consuming. A medium level of speed of delivery may therefore be guaranteed (score: 1). As stores are closer to the final customer, the speed of delivery can be high when orders are prepared in the store (score: 2). Orders prepared from multiple stock locations need a long time to be consolidated and can therefore only provide a low relative delivery speed (score: 0). Ease of access and return of products Customers expect easy access to the retailer to either pick up the products ordered or to return or exchange a product in case of dissatisfaction. Home delivery from a central warehouse may make returns and exchange difficult for customers (score: 0), who have to use a postal service for returning the products (except in the case the retailer owns a quite dense network of stores and the store channel is fully integrated to handle returns and exchange from the online channel). Order preparation in the store is usually done in the store closest to the customer. This store is therefore easy to access for the customer (score: 2). Multi-stock preparation can only provide a medium level of ease of access and return (score: 1). Although the customer can return a product to the closest store, exchange can be impossible if the products to be exchanged are only available at another stock location. Assurance The customers expect competent and friendly personnel in the store for buying, picking up their orders, return or exchange. The same expectation may apply for the personnel on telephone customer hotlines or the delivery personnel. These functions being very often outsourced (especially the latter), retailers have to focus on their own sales personnel and make sure they provide the level of assurance the customer is expecting by providing the same information to all channels. The assurance of the personnel can only be provided in the case of order pick-up of the customer in the stores (score: 2; other option scores: 0).

18


Stock efficiency For the retailer, the efficiency of stock management is crucial for numerous reasons. Stock keeping and monitoring of stock levels, balancing between different stock locations and the forecasting of sales and product reorders influence the availability of the products and therefore the sales. Efficient stock handling is as well an important cost element. Central stock management can provide a high level of efficiency (score: 2), whereas a store can only provide a medium efficiency (score: 1) because of the low number of products per article and the continuous stock movement due to customers purchases out of the shelves. The handling of multiple stocks in the fulfillment process is the least efficient (score: 0). Picking efficiency The picking of products out of stock to prepare the customer's order is the key process of the overall fulfillment process, representing a very important part of the warehouse operating costs. This process is nevertheless transparent to the customer. A central warehouse can be organized in a way to optimize the picking process by planning and reducing the way picking operators have to go during the process (score: 2). Depending on the volumes and the type of articles, it is possible to split the stock for whole cartons or open cartons picking or separate zones for stocking and picking. This optimization cannot be provided in a store or in multiple stock sites (scores: 0). Stores' layouts and product locations are planned to make the customer spend the maximum time in the store and not to optimize the picking (de Koster, 2002). Delivery cost efficiency Customers expect to pay the lowest extra cost possible for the delivery of their online orders and the retailer has to reduce the overall cost of the delivery. Delivery of orders prepared in a central warehouse has always to include the transport to the customer's home and therefore only achieve a medium efficiency (score: 1) compared to the delivery from a store (score: 2), where the distance to the final customer is short. Pick-up by the customer from the store where the order was prepared is the less costly way of delivery for the retailer. Any delivery from an order prepared in multiple stock locations (for home delivery or to a store for pick-up) is least efficient (score: 0), as the consolidation process of the order implies multiple costs of handling and transportation.

19


Channel organization efficiency As stated above, customers want to be able to use different channels simultaneously. A retailer has therefore to manage all the channels in an integrated manner. Preparing orders for home delivery from a central warehouse is the simplest and most efficient organization (score: 2). Order preparation in the store involves the online and the store channel in the fulfillment process, therefore adding a level of complexity (score: 1). Multi-stock preparation adds further complexity, being therefore the most difficult solution to achieve an efficient channel organization (score: 0). Fulfillment infrastructure costs The fulfillment process may generate important investment costs, when a specific site is built as the stock and the distribution center, or if the design and the organization of an existing facility have to be modified. Renting or outsourcing this infrastructure generates additional costs as well. A central warehouse order preparation generates a high infrastructure cost (score: 0), whereas the store preparation generates the lowest additional infrastructure costs (score: 2). Order preparation from multiple stocks may generate medium additional infrastructure costs (score: 1) to organize the consolidation of the orders. Return handling efficiency Customers expect their returns to be processed quickly, to receive an exchange product or to get reimbursed the purchasing price. The retailer must therefore provide a quick and efficient return handling process. A central warehouse enables creation of optimized return processing (score: 2) to check the returns, repack and restock them and send the exchange items to the customers. Multi-stock fulfillment can only provide a lower rate of return efficiency (score: 1), as the returned items arrive in different locations and the exchange items may only be available on different stock locations. For a store it is very difficult to offer an optimized return process (score: 0), especially when returned products are not part of the store's product offer.

We can therefore summarize the impacts of the different fulfillment systems on the different parameters as follows (Figure 3):

20


Figure 3: The Analysis Framework

Home Delivery Fulfillment system

Customer Store Pick-up 2.1 2.2

1.1

1.2

1.3

Prepared in central warehouse

Prepared in store

Prepared in multiple stocks

Prepared in store

Prepared in multiple stocks

High (2)

Medium (1)

Low (0)

Medium (1)

Low (0)

High (2)

Medium (1)

Low (0)

Medium (1)

Low (0)

Medium (1)

Low (0)

High (2)

Low (0)

High (2)

Medium (1)

Low (0)

High (2)

Low (0)

High (2)

Customer Expectations

Quality and quantity of information Reliability and respect of commitments Offer Product availability Speed of delivery Ease of access and return of products Assurance

Medium (1)

High (2)

Low (0)

High (2)

Low (0)

Low (0)

High (2)

Medium (1)

High (2)

Medium (1)

Low (0)

Low (0)

Low (0)

High (2)

High (2)

Economic performance

Stock efficiency Picking efficiency Delivery cost efficiency Channel organization efficiency Fulfillment infrastructure costs efficiency Return handling efficiency Total Score Customer expectations Total Score Economic performance Total

High (2)

Medium (1)

Low (0)

Medium (1)

Low (0)

High (2)

Low (0)

Low (0)

Low (0)

Low (0)

Medium (1)

High (2)

Low (0)

High (2)

Low (0)

High (2)

Medium (1)

Low (0)

Medium (1)

Low (0)

Low (0)

High (2)

Medium (1)

High (2)

Medium (1)

High (2)

Low (0)

Medium (1)

Low (0)

Medium (1)

7

6

5

8

7

9

6

2

6

2

16

12

7

14

9

This indicates several results: - Home delivery from a central warehouse (System 1.1) is the fulfillment system best meeting the combination of the two dimensions customer expectations and economic performance, providing a total score of 16. This system provides the highest score in economic performance (9) and the second highest score in customer expectations (7). - Customer pick-up after preparation in the store (system 2.1) provides the highest score in customer expectation (8) and a medium score in economic performance (6). The overall performance of this system is second best (total score of 14). - Home delivery after store preparation (system 1.2) provides medium scores in customer expectations (6) and economic performance (6), ranking third best in the overall score (12). - Any fulfillment system implying the order preparation out of multiple stocks appears to be less efficient: If the order is picked up by the customer (system 2.2), the total score is 9, if the 21


order is delivered at home (system 1.3), the total score is only 7. In terms of economic performance, these two systems provide the lowest score (2). - Economic performances of order preparation in the store or from multiple stocks are identical for both home delivery or in-store pick up by the customer. The difference in customer expectations (difference of 2) results from the possibility to have a personal contact with the customer, therefore responding to the customer expectation of 'Assurance', which can only be provided in a store.

DISCUSSION, MANAGERIAL IMPLICATIONS AND RESEARCH AVENUES The qualitative analysis of the different fulfillment systems and their impact on customer expectations and economic performance reveals that different fulfillment systems can provide high overall performance levels in terms of customer expectations and economic performance. The framework shows that orders for home delivery prepared in a central warehouse present only a slightly higher overall performance than an order prepared in a store for customer store-pick-up. A retailer has to assess the importance of each of the analyzed parameters for his specific retail mix and to optimize the trade-offs between channel fulfillment synergies and specific needs on different channels. Depending on the product mix between food and non-food products, the width of product assortment and the characteristics of the existing store network, specific solutions may be more appropriate than others to fit with a specific business model. Whereas proximity to customers is best achieved through a store network, stores add complexity on different levels in the fulfillment process which affect timing, costs and reliability. Multi-stock consolidation seems to be always by far the less efficient model, regarding the customer expectations and the economic performance. This corresponds to the three initially mentioned multi-channel retailers: Fnac abandoned the multi-stock preparation and completely integrated their fulfillment for the different channels, and neither Tesco nor Monoprix apply a multi-stock solution, whilst optimizing the in-store order preparation.

22


Our proposed framework could prove useful for multi-channel retailers in various ways: It provides a tool to evaluate the chosen fulfillment system and therefore identifying improvement potentials on specific parameters. It could also provide helpful to measure customer satisfaction. Moreover, it should clearly be useful in anticipating the needs for fulfillment systems when planning to implement a multichannel strategy. The proposed qualitative framework should be further validated by conducting supplementary empirical analysis, although it may be difficult to compare economic performances and customer expectations of different fulfillment systems in a same company. Another difficulty may be to isolate the impacts of the supply chain from other impacts (marketing, pricing, product policy etc.). A limitation of our framework is the fact that it does not assess the interdependencies between the different dependent parameters, but only focuses on one major impact at a time. A first step to further back this framework could be to conduct a dynamic analysis through a simulation, testing values for the different parameters of the analysis framework. Research could aim to find evidence if there is an 'optimum' fulfillment system for every multi-channel retailer. Such an optimal model could be defined depending on different independent variables like product assortment, average number of items per shopping basket, value of products, product characteristics, degree of channel integration etc. This would also include an analysis of the trade-off between cost- and customer service level optimization. Such a research could as well reveal that there is no optimum model, but just a set of independent and non stabilized models evolving quickly as the retailers' strategy mix evolves.

ACKNOWLEDGEMENT

The work on this paper was partially supported by the Innovation and Regulation Chair of Orange, Ecole Polytechnique, Telecom Paris.

23


REFERENCES Agatz N., Fleischmann M., van Nunen J. (2008), E-fulfillment and Multi-Channel Distribution – A Review, European Journal of Operational Research, 187, pp. 339-356. Avery J., Steenburgh T.J., Deighton J., Caravella M. (2009), Adding Bricks to Clicks: The Contingencies Driving Cannibalization and Complementarity in Multichannel Retailing, Harvard Business School, Working Paper 07-043, February 2009. Benghozi P.-J. (2001), Relations interentreprises et nouveaux modèles d'affaires, Revue économique, Vol. 52, No. hors série, pp. 165-190. Berman B., Evans J.R. (2010), Retail Management – A Strategic Approach, 11th edition, Pearson Prentice Hall (11th edition). Berman B., Thelen S. (2004), A Guide to Developing and Managing a Well-Integrated Multi-Channel Retail Strategy, International Journal of Retail & Distribution Management, Vol.32, N°3, pp. 147-156. Bressolles G., (2006), La qualité de service électronique: NetQu@l Proposition d'une échelle de mesure appliqué aux sites marchands et effets modérateurs, Recherche et Applications en Marketing, Septembre 2006; Vol. 21, N°3, pp. 19-45. Brown J. (2009), How Satisfied are French Consumers with Online Customer Experience? – Executive Summary, Forrester Research, Mai 2008. De Koster R. (2002), The Logistics behind the Enter Click, in Klose A., Speranza M.G., van Wassenhove L.N. (Eds.), Quantitative Approaches to Distribution Logistics & Supply Chain Management, Springer, Berlin, pp. 131-148. Dioux J. / Dupuis M. (2009), La Distribution – Stratégies des Réseaux et Management des Enseignes, Pearson Education France, Paris (2e édition). Durand B. (2008), Les modèles logistiques du "B to C" en France, 7e journée nantaise de recherche sur le e-marketing, 2008. Falk T., Schepers J., Hammerschmidt M., Bauer H. (2007), Identifying Cross-Channel Dissynergies for Multichannel Service Providers, Journal of Service Research, Vol. 10, N°2, pp. 143-160. Heinemann G. (2008), Multi-Channel-Handel – Erfolgsfaktoren und Best Practices, Gabler, Wiesbaden (2. Auflage). 24


Heinemann G. (2009), Verkauf auf allen Kanälen – Multi-Channel-Systeme erfolgsorientiert ausrichten, Marketing Review St. Gallen, 4-2009, pp. 46-51. Min H., Ko H.J., Ko C.S. (2006), A Genetic Algorithm Approach to Developing the MultiEchelon Reverse Logistics Network for Product Returns, Omega, N°34, pp. 56-69. Parasuraman A., Zeithaml V.A., Berry L.L. (1988), SERVQUAL: A Multiple-Item Scale for Measuring Consumer Perceptions of Service Quality, Journal of Retailing, Vol. 64, N°1, pp. 12-40. Parasuraman A., Zeithaml V.A, Malhotra A. (2005), E-S-Qual: A Multi-Item Scale for Assessing Electronic Service Quality. Journal of Service Research, Vol. 7, N°3, pp. 213233. Schobesberger A. (2007), Multichannel-Retailing im Einzelhandel – Entwicklung, Motivation, Einflussfaktoren, VDM Verlag Dr. Müller, Saarbrücken. Schröder H. (2005), Multichannel-Retailing – Marketing in Mehrkanalsystemen des Einzelhandels, Springer, Berlin & Heidelberg. Smith D., Sparks L. (2009), Tesco's supply chain management, in Fernie J., Sparks L. (Eds.), Logistics & Retail Management – Emerging issues and new challenges in the retail supply chain, Kogan Page, London & Philadelphia (3rd edition), pp.143-171. Swaminathan J.M., Tayur S.R. (2003), Models for Supply Chains in E-Business, Management Science, Vol. 49, N°10, pp. 1387-1406. Turban E., Lee J.K., King D., Liang T. (2009), Electronic Commerce, Pearson Prentice Hall (6th edition). Vanheems R. (2009), Distribution Multicanal – Pourquoi les clients mixtes doivent faire l'objet d'une attention particulière ?, Décisions Marketing, N°55, pp. 41-52. Venkatesan R., Kumar V., Ravishanker N. (2007), Multichannel Shopping: Causes and Consequences, Journal of Marketing, Vol. 71, pp. 114-132. Wirtz B. (2008) Multi-Channel Marketing – Grundlagen-Instrumente-Prozesse, Gabler, Wiesbaden.

25



How to compete by taking advantage of logistical collaborations? The LSP at the heart of coopetitive strategies François Fulconis Université d’Avignon & des Pays de Vaucluse, CRET-LOG, France [email protected] Virginie Hiesse Université Montpellier I, ERFI, France [email protected] Gilles Paché Université de la Méditerranée (Aix-Marseille II), CRET-LOG, France [email protected]

Abstract The subject of coopetitive strategies has been increasingly studied since the end of the 1990s in academic literature on strategic management, offering an in-depth renewal of the way of looking at the evolution of exchange relations between firms, and emphasizing that neither pure cooperation nor pure competition are satisfying explanatory models of the reality of business. In fact, cooperation and competition blend to give rise to a novel model of value creation: coopetition. The paper makes use of this conceptual framework to study the role of logistics service providers (LSPs) in the operation of modern supply chains. LSPs appear to be the catalysts of coopetitive strategies, particularly when they manage modular platforms. An exploratory field study conducted in France has identified three LSP archetypes in the dynamics of coopetitive supply chains: LSP as caretaker, LSP as lead operator and LSP as architect. Key words: Coopetition, France, Logistics industry, LSP, Strategy, Supply chain.

1


INTRODUCTION

One bleak and hazy morning, a group of lecturers and students were visiting a warehouse located near Marseilles, in the South of France, specialized in the stock management of convenience goods for hypermarkets and supermarkets. A common event since large retailers decided to have full control of the stocking up of their stores, rather than leaving the task to manufacturers and/or wholesalers. But a surprise was awaiting the visitors when the warehouse manager explained that the major two French large retailers, Alpha and Beta, in direct competition on the market for thirty years, had decided to share the logistical resources on the site to improve the performance of their respective supply chains. Aware of the difficulties of implementing this strategy of pooling, the two large retailers called in a logistics service provider (LSP) who plays the role of third party player, and developed particularly sophisticated technical solutions in order management and picking operations. In brief, Alpha and Beta remain fierce competitors in attracting consumers in their stores; they conduct a commercial war where every marketing trick is allowed, but they cooperate strongly in matters of construction and management of supply chains. Two years after the beginning of the pooling process, the warehouse manager concluded that delivery costs have decreased by nearly 12 per cent and customer service increased by 20 per cent for both firms! Despite its apparently paradoxical aspect, a situation mixing competition and cooperation has become quite common, and undoubtedly opens renewed outlooks for logistical thought. The dynamics of supply chain development have been increasingly analyzed over the past fifteen years. At first, the operating procedures of logistical systems associating a set of stakeholders (manufacturers, large retailers, suppliers, etc.) were studied, as well as the collaborative monitoring tools used to increase their reactivity. Now, questions of a more strategic nature about value creation and distribution are taking over. Most research papers, following Christopher (2010), agree in stating that the dominant competitive process is based on a confrontation between supply chains; it is however necessary to point out that business facts lead to a coexistence between competition and cooperation within supply chains, as the example of Alpha and Beta shows. They correspond to the notion of coopetition as defined by Brandenburger & Nalebuff (1996), i.e. that competitors in a market, be they suppliers, manufacturers or large retailers, are sometimes well advised to adopt cooperative strategies to improve their respective positions, in matters of procurement or logistics management for example. In many industrial sectors, it is frequently noted that companies cooperate in the

2


assembling activity by implementing shared modular platforms while competition rages in the marketing activities. On the logistics side, collaboration (as a cooperative strategy) appears to be a pooling of resources, a must for reducing the operating costs of supply chains. We deal with the subject from the assumption that LSPs play the role of coopetitive strategy catalysts in supply chains, relying on their expertise in flow management for manufacturers and/or large retailers. As early as the 1980s, LSPs positioned themselves at the interface of several rival supply chains to better manage the required resources and competences according to a pooling pattern. Gradually, French companies such as the Geodis Group and Norbert Dentressangle Group became supply chain lead operators by implementing collective solutions improving each of their customers’ logistical performance. The most significant illustration of this phenomenon lies in the organization of physical distribution activities, with a sharing of stock, transport and even co-packing management resources. These activities are particularly sensitive to scale and learning economies. LSPs enabled some of their customers who were in direct competition to reach the required critical size to develop a sustainable competitive advantage (the case of small canned food manufacturers at the beginning of the 1980s). The paper examines this research subject in reference to an exploratory field study conducted in France, seeming to prove that in the end, LSPs can play three different roles in the management of coopetitive supply chains.

1: LSP–A COOPETITIVE PERSPECTIVE

Coopetition has not produced an in-depth theoretical reflection in matters of management of supply chains. Analyses remain based on inter-firm cooperation or on inter-firm competition and assess the advantages and drawbacks of each option. Of course, there are some studies under way as illustrated by Kotzab & Teller’s (2003) contribution on value-adding partnerships in the grocery sector, Lee & Song’s (2007) contribution on the strategies of port operators, or Spens & Kovács’s (2007) contribution on humanitarian relief supply chains, but no real research agenda on the subject has emerged. And yet, it seems particularly pertinent to refer to the notion of coopetition for a better understanding of the behaviours of LSPs developing efficient solutions for a number of directly competitive customers, while pooling logistical resources for them. The case of modular platforms, implemented by some LSPs, highlights the relevance of approaching supply chain dynamics through coopetition.

3


1.1: An enlarged services approach

In the last fifteen years, a significant academic literature on marketing, supply chain management and strategy has been trying to define the profile of the logistics industry in reference to operations carried out by LSPs, with various degrees of complexity depending on the outsourcing agreements signed with manufacturers and/or large retailers (Andersson & Norrman, 2002; Roques & Michrafy, 2003; Marasco, 2008; Fabbe-Costes et al., 2009; Fulconis et al., 2010). It is obviously impossible to sum up their contributions and limits here. We will simply mention a recent survey conducted in France by the Observatoire de la Prestation Logistique, the interest of which is to divide the service offer into three distinct parts: a first feature called “Core business” (transport, warehousing, order preparation, etc.), a second feature called “Additional customer services” (after-sales service, invoicing, etc.) and a third feature called “New professions” (co-packing, co-manufacturing, etc.) (Roques & Michrafy, 2003). Although the LSP trade is generally defined as exercising logistical activities on behalf of manufacturers or large retailers, it varies greatly both in terms of size and the services provided. Dornier & Fender (2007, p. 356) mentioned that a large number of LSPs are “characterized by a bipolar structure including micro-enterprises with a few employees and major firms with sales turnovers amounting to several billions of euros”. As to the diversity of services offered, the Accenture consultancy registered “appellations” to identify the different types of LSPs on this criterion. In addition to the activity of 1PL (first party logistics) exclusively involving transport or warehousing contracting, and the activity of 2PL (second party logistics) involving transport and warehousing outsourcing only, there are two other categories of now common activities: 3PL (third party logistics) and 4PL (fourth party logistics). Most 3PL come from the road haulage sector and ensure the simple execution of physical operations associated with product transport, handling and storage, and even the management of industrial or commercial operations (mass customization), administrative operations (invoicing) and information operations (tracking-tracing) (Selviaridis & Spring, 2007). Unlike 3PL who own their transport and warehousing means, the objective of 4PL is to design and sell tailored logistical solutions (particularly information systems) by creating a network of competences associating for example hauliers, warehousers and sub-contractors (see Figure 1).

4


TRANSPORT

CONVENTIONAL LSP

COMPANY

−

4PL

Value added services

Transport management

Warehousing and distribution

Final assembly

+ Coordination of supply chain operations

Figure 1: From transport company to 4PL

While the French case shows an obvious widening of the LSPs’ service offers, it is far from being isolated and unrepresentative of the global trend. Many LSPs in Northern Europe, the United States and Asia have followed the same reasoning. Originally oriented toward transport and warehousing activities, the scope of their service offers now focuses on the management of many complex interfaces with the customers in the manufacturing and retailing industry. What is then essential for LSPs is to act as “go-betweens”, in other words establishing contacts between supply chain members, without LSPs necessarily owning material assets (Saglietto et al., 2007). The mediator function of LSPs is clearly illustrated by Carbone (2004). Drawing his inspiration from Chow & Gritta’s (2002) contribution, Carbone (2004) showed how some LSPs adopt a non asset based model that privileges flow coordination competencies rather than heavy investments in transport and logistics assets as per the asset based model. Interface management takes precedence over the control of the different elements in the offer. Of course, LSPs ensuring a mediator function remained a minority in Carbone’s (2004) sample, but they represent an emerging group not to be neglected by researchers. All the more so as other works present converging results in the reconfiguration of LSPS’ value chains. For example, Lai (2004) conducted a survey of LSPs in Hong-Kong to assess their service capabilities in two types of logistical services, in complement to conventional freight forwarding service (FFD): -

Value-added logistical services (VAL): assembling, packaging & labelling, purchasing & procurement, cross-docking, customer-specific label printing, warehousing, etc.

-

Technology-enabled logistical services (TEL): information systems management, tracking and tracing.

The field study concluded, from a cluster analysis, that there were four types of LSPs, with 27.1 per cent labelled as “Full service providers”, because their level of capability is high in each logistical service (FFD, VAL and TEL). The study also emphasizes the presence of an 5


unusual group of LSPs, called “Nichers”, who as a priority have developed capabilities in the value-added and technology-enabled logistical services (VAL and TEL), to the detriment of conventional freight forward service (FFD). Finally, LSPs centered on basic transport operations now only account for a quarter of the sector. The large number of LSPs who have developed new value-added services, as shown in Lai’s (2004) field study, shows the magnitude of logistical mutations in some South-Eastern Asian countries, that are probably occurring at a greater speed than in Northern Europe. This type of mutation is not due to chance. It simply corresponds to the fact that LSPs have known how to progressively widen their service offer to answer the expectations of manufacturers and large retailers who have also progressively abandoned many logistical activities to better dedicate themselves to the management of their core business. The case study of Jager et al. (2009) on the Swedish furniture industry revealed that the case company has evolved from being a standard 3PL provider to a service developer; in addition to traditional services, it provides more value-added services. These services involve a set of more standardized activities that can be combined according to each customer’s wishes and requirements. Among the most radical developments, we should mention that this LSP took over modular platform operation activities. Such activities raise formidable challenges to the way of considering value creation and distribution modes within supply chains (Fulconis & Paché, 2005). An LSP operating a modular platform mobilizes increasingly sophisticated resources and competences naturally leading it to offer its services to a growing number of customers. Customers also find it interesting as the LSP produces such significant economies of scale and scope that its offer is far more attractive than an internal management of operations. The history of the logistics industry shows how activities of promotional packs then of co-packing were transferred from manufacturers and large retailers to LSP warehouses. If we start from the principle that LSPs, through their modular platforms, are at the interface between several supply chains in ensuring high-value operations for each, essential for personalizing finished products, the question of which strategic dynamics are used becomes increasingly important: how to articulate cooperation and competition harmoniously? In modular platforms, LSPs share resources for several customers who are then in a situation of “indirect” cooperation in the assembling activity of modules for example. But these same customers will obviously be competitors in markets in capturing final customers with their offer of customized products. This is the specific configuration of the coopetition strategy,

6


formalized for the first time by Brandenburger & Nalebuff (1996). But more recent research on SCM, particularly in the North American context, is unwilling to integrate the dualism of the cooperation/competition paradigms and defend a purely normative view: collaborative practices are the only possible solution for firms involved in the operation of a supply chain, the sole stable state to reach to bring maximum value to final customers (Fawcett et al., 2008). So that reflection on an alternative paradigm close to the reality of the game between supply chain members is all the more stimulating.

1.2: Fundamentals of coopetition

Coopetition describes the fact that in our current business environment, firms can create more value and develop in markets if and only if they work together rather than on their own (Brandenburger & Nalebuff, 1996). They remain competitors in access to rare resources, but their interest lies in knowing how to cooperate judiciously for more efficient uses. The R&D process is an excellent case study as firms “may display cooperative interest structures at the time of co-creating value through an R&D project, but they may also undergo competitive pressures at the time of capturing the value created” (Cassiman et al., 2009, p. 217). The approach supplies a sort of diagnostic tool for clarifying situations. For example, it proves essential to know whether a firm will be confronted with a direct competitor or, in contrast, with a possible complementor who will bring additional resources, indispensable for enhancing the value of its offer. According to Brandenburger & Nalebuff (1996), to describe coopetition, the five following factors must be taken into account: -

the opposing players, by assessing their potential complementarities;

-

the different added values created by each player;

-

the rules of the game formalized between the players to organize the exchange;

-

the tactics used by each party, particularly to change the rules of the game;

-

the fields of tactics exercised within value networks (commercial, industrial, logistical, etc.).

The idea that cooperation and competition are finally two simultaneous and logical ways of acting rather than exclusive ones does not present any remarkable originality. The business world offers numerous examples in which competitors have managed to find a way to agree to develop a project or formalize a shared standard to facilitate the penetration of a new product onto the market. What is more original, on the other hand, is to consider the situation

7


of coopetition in terms of a sustainable strategic balance creating competitive advantage: a balance in which players no longer fear the reappearance of an opportunistic behaviour from a partner skilled in taking advantage of cooperation by later starting hostilities once having garnered the benefits of the cooperation phase. This led Bengtsson & Kock (1999) to detail the conditions required to implement a coopetition relation mode: competitors must both occupy a relatively strong position in the sector of activity (thus excluding any alliance of an asymmetric nature), and the requirement for outside resources must be high (thus excluding a structurally unbalanced situation of resource dependence). Aren’t alliances between competitors as part of a coopetitive strategy hazardous, by forcing partners to disclose knowledge and competences to a competitor? This is a conventional position held in the literature, but Ritala (2009) emphasized that coopetition also offers significant value creation opportunities. These distinctive risks and benefits arise from the fact that the collaborating firms are rivals in the end-product and strategic resource markets. Analyzing the impact of market rivalry on the basis of value creation in interfirm alliances, Ritala’s (2009) contribution showed that coopetition is not risky or beneficial by definition. In particular, the way that the competing firms design and manage the alliance with respect to market rivalry and their common and specialized knowledge and competences actually determine how the benefits and risks in such a relationship are structured. A coopetitive strategy has to be studied in a given context, as a function of the structure of markets and relations between players; in some cases, when the benefits of coopetition are high, to choose cooperation or competition will be a major strategic mistake. From the above reasoning, it appears that a coopetitive strategy is not simply a transitory step in an evolutionary path between competition and cooperation, or between cooperation and competition. It must be in contrast considered as a real end objective for players, who accordingly formalize stable rules of the game. Without going as far as saying that coopetition is the new managerial paradigm, it is an organized system in which players “interact on the basis of a partial congruence of interests and objectives” (Dagnino et al., 2007, p. 95). For these authors, the key element of this strategy is the identification of interdependences in the process of value creation, in the distribution of mutual benefits and in the detailed identification of convergent interests between players. This naturally leads to a research programme exploring, among other subjects, the intrinsic performance of a coopetitive relation mode in comparison with competition or cooperation. Questions on the pertinence of a coopetitive strategy within supply chains are already being introduced by some firms. They

8


oppose the dominant perspective in strategic management, an analysis of the supply chain environment mainly from the notion of hypercompetition (Kotzab et al., 2009).

1.3: Coopetition, mass customization and modular platforms

Let us go back to the case of mass customization and of the operation of a system of modular platforms by LSPs. The framework proposed by Padula & Dagnino (2007) may be used with some relevance. LSPs offer an attractive solution, both operationally and economically, allowing manufacturers and large retailers to customize their offer at an attractive cost. They accept the cooperative game on the modular platform as far as their leeway remains intact in their commercial fight to capture customers, particularly through an aggressive marketing action. Coopetition here is structurally complementary and imposed by the imperative of a low price strategy, according to Depeyre & Dumez’s (2009) analysis. But any asymmetry in terms of market power can threaten the survival of the coopetitive network in the end. The dominated supply chain member will wonder whether it is in its interest to go on playing, in other words to share resources in the modular platform (via the LSP) with other supply chain members who gain more than it in terms of market power. The debate is not new, it crops up from time to time in what Brandenburger & Nalebuff (1996) call the tactics exercise field (cooperative, competitive) within value networks built along coopetitive reasoning. Baumard (2009) reminded us of the main issue: to what extent will competitors accept sharing the exploitation and/or exploration of critical assets with other firms? The coopetitive dynamics raise the question of adapting the innovation strategies of individual players that can enable them to maintain their place in the coopetitive game without losing their individual capacity for innovation as a source of competitive advantage. It is obvious that a modular platform is a place of convergence of competences for the different partners, particularly at the level of the layout of each module, in other words in the management of interfaces. When the management of interfaces is considered as a critical asset for each partner, the risk of dilution of their competitive advantage becomes a reality. This could lead them to brutally withdraw from the modular platform, even at the price of a loss of economies of scale. If any system of players with cooperative views is threatened by the intrusion of competition, with the risk of unsettling it irreversibly, the part that a third party could hold in the perpetuation of a coopetitive relation mode becomes a stimulating research path.

9


Unsurprisingly, this subject is one of the major points in the research agenda suggested by Walley (2007). The advantage of a third party lies in not being involved in the organizational system, moves for power and culture of each player; in contrast, a third party is a “referee” capable of proposing original solutions to potential conflicts of interest. This is the function that an LSP can fill in the operation of a modular platform and in the associated information exchange. This last point is crucial as the LSP occupies the position of hub in an information network linked to the monitoring of materials and product flows in the supply chain. The capacity of data processing and memorization is a significant element for ensuring the adequate reactivity of firms faced with the high diversity of references required by consumers. Coopetition finds a favoured field of application in the operation of current supply chains especially as the technical solutions for optimizing flow monitoring operations naturally require rapprochements between direct competitors. The improved management of delivery rounds, to avoid empty return trips for example, led manufacturers in the same sector to share their transport resources a long time ago; this issue is now even more topical in a context of sustainable development requiring a reasoned use of logistical resources (Monnet, 2007). Considering the central place of LSPs in supply chains, particularly when ensuring an assembling activity within modular platforms, they become key players in an extremely sophisticated logistical coordination at technological level. This fact is now acknowledged in the academic literature, but it must be admitted that the ability of LSPs to favour the implementation of a coopetitive relation mode has not often been studied. An exploratory research conducted in France suggests a few paths for reflection in this direction.

2: IDENTIFICATION OF THREE COOPETITIVE ARCHETYPES

For several years, manufacturers in direct competition in terms of developing consumer loyalty, but working for the same large retailer, have chosen a pooling of logistical resources. The experience conducted by the three direct competitors Henkel, Colgate and ReckittBenckiser is well-known: since 2005, the three manufacturers have shared the same warehouse to increase delivery frequency to stores without increasing transport costs. Basing themselves on an exploratory field study, we tried to determine whether other similar experiments were under way, and what place LSPs occupy in them. The exploratory field study led to a first assessment of cases of pooling between competitive manufacturers. They

10


are finally far more numerous than those reported by the trade press, but mostly remain confidential.

2.1: A network approach of LSP coopetitive strategy

The network approach initiated by Snow et al. (1992) represents a pertinent conceptual framework for understanding developments in the logistics industry. This approach suggests a representation of inter-organizational relationship dynamics, which, while being sometimes simplistic, remains very instructive as to the part that LSPs can play in it. According to Snow et al. (1992), inter-organizational relations lead to the creation of network architectures with a decision centre, a firm at the heart of a relationship and information network, weaving close links with suppliers, customers, LSPs, competitors, etc. It becomes apparent that such organizational forms generally include two elements: -

the network core, usually occupied by a major firm, but not necessarily –a PME can fulfill the role–, with a variety of designations (broker, hub firm, or strategic centre);

-

a relational area, including satellite firms, located around the network core, generally specialized in the manufacturing of one module, the physical distribution of one product, or the performance of logistical services.

In this approach, the LSP’s job is to put at the disposal of the firm occupying the strategic centre a web of logistical modules (transport module, storage module, order preparation module), while suppliers and producers put at the disposal of the strategic centre components and sub-assemblies modules, and distributor, sales and marketing modules. The network architecture is monitored by the hub firm responsible for assembling the various modules. In this quality of satellite, the LSP is here in a “peripheral” situation. But although the outsourcing phenomenon is now widely used by firms, its application in the field of logistics is more recent. New methods of flow management, new functions, but also new professions practised by value-added LSP (3PL) and increasingly by dematerialized LSP (4PL), are implemented along the supply chains. LSPs integrate coordination competences and have warehouses increasingly looking like modular platforms. Eventually, they simply monitor the activities of the network members thanks to their efficient control of inter-organizational information systems. 3PL and 4PL appear to be key players in supply chains constructed between manufacturers on one side and large retailers on the other. In reality, the planning of logistical activities is done

11


essentially through a total control of information about physical flows of components, subassemblies and finished goods; this control enables LSPs to play a central role in the monitoring of network architectures. We can see that these new-generation LSPs are increasingly used in the USA in many sectors of activity, and they appear to be developing fast in Belgium, Germany and the Netherlands. This is an opportunity to try and better define the LSP’s change of role in the logistics industry, particularly as potential brokers within a network architecture. This function of potential broker is particularly interesting as it allows an LSP to stand at the interconnection of several supply chains, by working simultaneously for firms in direct marketing competition, but that will cooperate logistically, in operational management.

2.2: Research methodology

Taking into account the novelty of the coopetition phenomenon in logistics and the lack of works on it, we opted for an exploratory field study conducted in two steps. First, a review of secondary data from the specialized professional press helped in identifying the existence of new logistical patterns in the French food retailing industry, and also in contractual networks. Since the 2000s, a number of initiatives for pooling resources and logistical activities have developed between manufacturers working for the food retailing industry; these initiatives lead LSPs to manage storage and transport means and also store supplying between direct competitors. Then, the exploratory field study was conducted in the convenience goods sector. It is based on fifteen semi-directed interviews, between May 2008 and September 2009, with LSPs and consultants involved in pooling approaches. The interviewed individuals were identified through the analysis of the secondary data, but for reasons associated with the confidential nature of the data collected, they did not wish their names, or the names of their companies to be quoted in the paper. Interviews were face-to-face, individual interviews, using an interview grid listing the following questions: the history of the pooling process, the general functioning of the logistical pattern implemented on the coopetition model, the nature of relationships between the different firms and the awareness of the pooling consequences. Each interview being synthesized, we continued with a conventional analysis of the subjects broached. Note that interviewees spoke rather freely of the organization methods and economic outlook of the new organizational architecture. But this was not the case when the

12


strategic outlook was mentioned as the individuals we met are reticent to give information on prospective subjects. This is explained by the sensitive nature of numerous pooling situations between manufacturers, which can also look like anti-competition practices in the eyes of the UE antitrust rules. However, the exploratory field study leads to an early assessment of the identified cases.

2.3: Main results

We wish to present a summary of the major findings on the part now played by LSPs as catalysts of coopetitive strategies in the supply chain. Using the Snow et al.’s (1992) terminology, our exploration identified three archetypes: LSPs as caretakers, LSPs as lead operators and LSPs as architects. Table 1 describes the processes and the major behavioural features. In most cases studied, the logistical resource pooling approach is designed by the competitive manufacturers in agreement with a large retailer; they look for an LSP who will be able – thanks to its competences and know-how – to play the part of caretaker and/or lead operator of coopetitive supply chains. More recently, 4PL type LSPs seem to have developed a new expertise enabling them to be the architects of coopetitive supply chains, but monitoring coopetition relations between competitive manufacturers. One or several LSPs seem to be able to play one or several roles in the same coopetitive network.

LSPs’ role

Process

Features

Caretaker

Exploitation of coopetition relations

Lead operator

Coordination of coopetition relations

Architect

Formalization of coopetition relations

- Performs a logistical activity - Creates communications and exchange areas - Advises, legitimizes the approach - Plays the part of third party and makes the network last - Coordinates objectives, processes, flows, systems and technologies - Implements dedicated tools and approaches - Optimizes logistical pooling - Starts and builds the value network - Catalyzes cooperation - Manages the arrival and departure of supply chain members

Table 1: LSPs’ roles in coopetitive strategies Source: Adapted from Hiesse (2009).

13


2.3.1: LSP as caretaker

LSPs ensure the operation of the coopetitive network by helping the implementation of the supply chain members’ strategy and the development and smooth course of exchanges between them. Conventional, 3PL-type LSPs, take charge of a support activity associated with the logistical pooling between competitors: warehouse management, transport management, etc. As such, LSPs may be called operation links in Zhang’s (2006) sense. LSPs also create shared communication and exchange areas facilitating the implementation of coopetitive strategies. They invite their customer manufacturers to information meetings on innovative approaches. The fact of describing logistical pooling approaches implemented by other manufacturers and the associated results gives them some legitimacy and reduces the psychological barriers created by the fact of working with competitors. LSPs also conduct opportunity surveys for one or several manufacturers to determine with which competitors it would be pertinent to form a pool; LSPs organize their meeting and help them to agree. Their in-depth knowledge of the players, their needs, their products and logistical issues enable them to present themselves as experts in supply chains. Finally, LSPs as caretakers act as third parties and favour the creation of trust relationships between competitive manufacturers. LSPs guarantee the adherence to rules fixed by competitors: organization (products stored in partitioned areas); confidentiality of strategic information; fair and equal treatment of partners. LSPs also play the part of filter in collecting all the information they need (strategic or not), and in presenting to each partner only the relevant corresponding logistical information. LSPs facilitate the issue and dissemination of some information and prevent access to other strategic information, a fact, according to Prévot (2007), representing a key factor in the successful monitoring of coopetition relations. LSPs can also ensure the management of conflicts between competitive partners as a kind of regulator. In their role of caretakers, LSPs seem to mobilize logistical competences and relationship competences, particularly in interface management.

2.3.2: LSP as lead operator

LSPs as lead operators have to suggest practical paths for improving the operational management of the coopetitive network designed by the supply chain members. LSPs are first of all responsible for taking into account all the objectives defined by their customers. From

14


there on, they coordinate flow management, the execution of logistical activities (transport, storage, procurement), the interoperability of the competitive manufacturers’ information systems and technologies. They have to organize and optimize the sharing of the partners’ resources. They are the key players of logistical pooling between manufacturers. LSPs serve as an interface between horizontal partners (manufacturers), at the same level in the supply chain, as well as at successive levels in the supply chain, between these manufacturers and large retailers. Logistical competence relies upon the control of information systems feeding the supply and physical distribution networks. LSPs participate in the designing and implementation of innovative tools and approaches (integrated packages, dedicated information systems, shared logistical patterns). As lead operators, they also have relationship competences to ensure interface management. They conduct monthly meetings with the partner manufacturers, facilitate negotiations, and build internal project teams who work at organizational interfaces (with manufacturers, large retailers, etc.). Taking coordination in charge is all the more important if the level of logistical pooling between competitors is high. Through the cases studied, we found that the period required for all involved players to find an agreement varies from one to two years. Once the agreement is formalized, several contracts are signed between the parties, in order to pilot their actions for a period of three to five years.

2.3.3: LSP as architect

LSPs as architects design the coopetitive network (i.e. its potential members), ensure negotiations about shared strategies and the objectives to select, and formalize logistical patterns. LSPs conduct surveys to determine whether competitive manufacturers had better form a pool. In general, one or several of these companies are already the LSP’s customers, and are even grouped in multi-customers sites. By pointing out the advantages of a logistical pooling approach to each member of the future pool, LSPs will then organize meetings and propose a shared logistical pattern. They rely on their expertise accumulated in the course of coopetition experiences where they are caretakers or lead operators, and on their recognized expertise as third parties between manufacturers and large retailers. In their role of architects, LSPs launch and build networks, catalyze cooperation between the supply chain members, and may also be called on to manage the arrival or departure of supply chain members. Thus, LSPs mobilize logistical, relationship and architectural competences.

15


Modular platforms illustrate the role of architect played by LSPs. They are based on the management of logistical interfaces whose control allows LSPs to later transform themselves into supply chain “maestri” (Arroyo-López & Bitran, 2007). In other words, logistical competence is an indispensable prerequisite, but a prerequisite only, to be able to control product-architecture. As a reminder, a product-architect, in a context of mass customization, defines and adheres to explicit rules in matters of module building and of interface management so as to put modules together (Baldwin & Clark, 2000). The logistical dimension is but one of the components of product-architecture, essential to ensure market entry, but without intrinsic value if industrial and commercial dimensions are not integrated. Could LSPs assimilate them? According to Sturgeon (2002), the central know-how required from any product-architect is to be able to quickly reconfigure interfaces between modules depending on developments in demand. The expertise accumulated by some LSPs in the last twenty years in supply chain management tends to indicate they can adopt such an approach in areas as distinct as micro-computing and household appliances.

CONCLUSION

To speak of supply chains today is to refer to a relationship system “through which organizations deliver goods and services to customers; this chain makes up a network of interconnected organizations with a shared purpose” (Samii, 2001, p. 6). Efforts to achieve the integrated management of supply chains, deemed more efficient for stakeholders, gave birth to an original approach whose purpose is to federate companies around joint objectives: SCM. SCM was widely advertised from the 1990s, and finds an obvious echo in applied research, as the hundreds of academic papers published every year on the subject show. Such a keen interest finds its origin in the desire of firms to react to customers’ demands almost in real time, by being able to maintain a favourable competitive position by regularly introducing new products in satisfying conditions of cost and service quality. LSPs now play an essential part in the operation of supply chains, and take over increasingly large intermediation tasks, from the running of conventional logistical operations to the running of postponement operations, and above all facilitate the pooling of logistical resources to the benefit of a network of interconnected supply chains. Competition and cooperation thus become the two indissociable faces of the coopetitive value creation process, and LSPs appear as key players as caretakers, lead operators or architects.

16


But as Luo (2007) pointed out, coopetition does not form a homogeneous relationship spectrum; its evolution is deeply influenced by the intensity of cooperation and competition at a given moment in the building, then the development of exchange relations between firms. Moreover, we can question whether coopetition itself should not be understood in a diachronic manner from the existence of strategic sequences (Dumez & Jeunemaître, 2006). In other words, competition and cooperation will sometimes follow each other in the same dimension for two firms A and B. If A and B use the same LSP in the running of operations of differentiation reported at time t by playing on resource pooling patterns, nothing allows us to state that at time t+1, the situation will remain identical, for example in case of radical modification of customers’ expectations, preferring A’s products rather than B’s products. LSPs will have to learn to manage such strategic switches by being able to rearrange postponement operations very quickly, or risk losing the accounts of both firms now in a competitive sequence after playing the cooperation game. We see that issues remain unexplored and justify continued research, by more closely associating strategic management and SCM. This effort is essential to achieve the universalistic objective of logistical thought: “How can we design and manage a supply chain, controlling its assets and uncertainties, to best meet the needs of the customers in a cost-effective manner?” (Ellram et al., 2004, p. 21).

REFERENCES

o

Andersson, D., Norrman, A. (2002), Procurement of logistics services–a minutes work or a multi-year project?, European Journal of Purchasing & Supply Management, Vol. 8, No. 1, pp. 3-14.

o

Arroyo-López, P., Bitran, G. (2007), Coordination of supply chain networks and the emergence of mini-maestros, MIT Sloan School Working Paper No. 4674-08, Cambridge (MA).

o

Baldwin, C., Clark, K. (2000), Design rules: the power of modularity, Cambridge (MA): MIT Press.

o

Baumard, P. (2009), An asymmetric perspective on coopetitive strategies, International Journal of Entrepreneurship and Small Business, Vol. 8, No. 1, pp. 6-22.

o

Bengtsson, M., Kock, S. (1999), Cooperation and competition in relationships between competitors in business networks, Journal of Business & Industrial Marketing, Vol. 14, No. 3, pp. 178-190. 17


o

Brandenburger, A., Nalebuff, B. (1996), Co-opetition: a revolution mindset that combines competition and cooperation, New York (NY): HarperCollins.

o

Carbone, V. (2004), Le rôle des prestataires logistiques en Europe: intégration des chaînes et alliances logistiques, Thèse de doctorat en Transport, Ecole Nationale des Ponts et Chaussées, June.

o

Cassiman, B., Di Guardo, M.-C., Valentini, G. (2009), Organising R&D projects to profit from innovation: insights from co-opetition, Long Range Planning, Vol. 42, No. 2, pp. 216-233.

o

Chow, G., Gritta, R. (2002), The North American logistics service industry, Proceedings of the 4th International Meeting for Research in Logistics, Lisbon, pp. 1-51 (CD-Rom).

o

Christopher, M. (2010), Logistics and supply chain management, Harlow: FT-Prentice Hall, 4th ed.

o

Dagnino, G.-B., Le Roy, F., Yami, S. (2007), La dynamique des stratégies de coopétition, Revue Française de Gestion, No. 176, pp. 87-98.

o

Depeyre, C., Dumez, H. (2009), Le concept de coopétition: quelques voies de recherche à partir d’une analyse de cas, Le Libellio d’Aegis, Vol. 4, No. 3, pp. 13-21.

o

Dornier, P.-P., Fender, M. (2007), La logistique globale et le supply chain management: enjeux, principes, exemples, Paris: Editions d’Organisation, 2nd ed.

o

Dumez, H., Jeunemaître, A. (2006), Etudier la combinaison affrontement et coopération: la notion de séquence stratégique multidimensionnelle, in Yami, S. & Le Roy, F. (Eds.), Stratégies collectives: rivaliser et coopérer avec ses concurrents, Caen: Editions Management & Société, pp. 37-47.

o

Ellram, L., Tate, W., Billington, C. (2004), Understanding and managing the services supply chain, Journal of Supply Chain Management, Vol. 40, No. 4, pp. 17-32.

o

Fabbe-Costes, N., Jahre, M., Roussat, C. (2009), Supply chain integration: the role of logistics service providers, International Journal of Productivity and Performance Management, Vol. 58, No. 1, pp. 71-91.

o

Fawcett, S., Magnan, G., McCarter, M. (2008), A three-stage implementation model for supply chain collaboration, Journal of Business Logistics, Vol. 29, No. 1, pp. 93-112.

18


o

Fulconis, F., Paché, G. (2005), Piloter des entreprises virtuelles: quel rôle pour les prestataires de services logistiques?, Revue Française de Gestion, No. 156, pp. 167-186.

o

Fulconis, F., Paché, G., Roveillo, G. (2010), La prestation logistique: origines, enjeux et perspectives, Caen: Editions Management & Société, forthcoming.

o

Hiesse, V. (2009), L’intermédiation du PSL dans les canaux de distribution: quels schémas logistiques émergents?, Logistique & Management, Vol. 17, No. 2, pp. 29-40.

o

Jager, K., Hilletofth, P., Ujvari, S. (2009), From standard 3PL provider to service developer: a case study from the Swedish furniture industry, World Review of Intermodal Transportation Research, Vol. 2, No. 4, pp. 376-390.

o

Kotzab, H., Teller, C. (2003), Value-adding partnerships and co-opetition models in the grocery industry, International Journal of Physical Distribution & Logistics Management, Vol. 33, No. 3, pp. 268-281.

o

Kotzab, H., Grant, D., Teller, C., Halldorsson, A. (2009), Supply chain management and hypercompetition, Logistics Research, Vol. 1, No. 1, pp. 5-13.

o

Lai, K. (2004), Service capability and performance of logistics service providers, Transportation Research Part E, Vol. 40, No. 5, pp. 385-399.

o

Lee, E.-S., Song, D.-W. (2007), Towards co-opetition in global logistics and supply chain: a conceptual framework for port operators, Proceedings of the of the 12th Logistics Research Network Conference, Kingston Upon Hull, pp. 218-223.

o

Luo, Y. (2007), A coopetition perspective of global competition, Journal of World Business, Vol. 42, No. 2, pp. 129-144.

o

Marasco, A. (2008), Third-party logistics: a literature review, International Journal of Production Economics, Vol. 113, No. 1, pp. 127-147.

o

Monnet, M. (2007), L’intermédiation du prestataire de services logistiques dans une “supply chain” en contexte de développement durable, Thèse de doctorat en Sciences de Gestion, Université de la Méditerranée (Aix-Marseille II), November.

o

Padula, G., Dagnino, G.-B. (2007), Untangling the rise of coopetition: the intrusion of competition in a cooperative game structure, International Studies of Management & Organization, Vol. 37, No. 2, pp. 32-52.

19


o

Prévot, F. (2007), Coopétition et management des compétences, Revue Française de Gestion, No. 176, pp. 183-202.

o

Ritala, P. (2009), Is coopetition different from cooperation? The impact of market rivalry on value creation in alliances, International Journal of Intellectual Property Management, Vol. 3, No. 1, pp. 39-55.

o

Roques, T., Michrafy, M. (2003), Logistics service providers in France–2002 survey: actors’ perceptions and changes in practice, Supply Chain Forum: An International Journal, Vol. 4, No. 2, pp. 34-52.

o

Saglietto, L., Fulconis, F., Paché, G. (2007), 4PL and models of strategic alignment–A research note, Networks and Communication Studies, Vol. 21, No. 3-4, pp. 395-406.

o

Samii, A.-K. (2001), Stratégies de service. E-business, supply chain, Paris: Dunod.

o

Selviaridis, K., Spring, M. (2007), Third party logistics: a literature review and research agenda, International Journal of Logistics Management, Vol. 18, No. 1, pp. 125-150.

o

Snow, C., Miles, R., Coleman, H. (1992), Managing 21st Century network organizations, Organizational Dynamics, Vol. 20, No. 3, pp. 5-20.

o

Spens, K., Kovács, G. (2007), Knowledge sharing in relief supply chains–The challenge of co-opetition, Proceedings of the of the 12th Logistics Research Network Conference, Kingston Upon Hull, pp. 609-614.

o

Sturgeon, T. (2002), Modular production networks: a new American model of industrial organization, Industrial and Corporate Change, Vol. 11, No. 3, pp. 451-496.

o

Walley, K. (2007), Coopetition: an introduction to the subject and an agenda for research, International Studies of Management & Organization, Vol. 37, No. 2, pp. 11-31.

o

Zhang, D. (2006), A network economic model for supply chain versus supply chain competition, Omega, Vol. 34, No. 3, pp. 283-295.

20



Bordeaux natural wines on the Japanese market: analysis of supply chain system' indolence Tatiana BOUZDINE-CHAMEEVA BEM Bordeaux Management School, France [email protected]

Marie NINOMIYA Faculty of Commerce, Fukuoka University, Japan [email protected]

Abstract Over the last few years, the interest towards natural wines has been increasing on this world wine market and in Japan in particular. The Bordeaux natural wines have been facing challenges on the market, partly due to the supply chain problems. Our work focuses on understanding the role and interactions of major actors on the Bordeaux wine scene and natural wines' distribution in Japan. The research methodology employs content analysis and critical incidences approach. A study is grounded on a series of interviews with winemakers, wine merchants, wine broker's agency, a union of natural wine producers and several natural wines' distributors in Japan. Our findings show the strong heterogeneity of the Bordeaux natural wines' production sector, considerable differences in supply chain solutions. The practical implementation of these results is discussed.

Key words: Wine supply chain, Natural wines, Distribution in Japan, Bordeaux wines.

1


INTRODUCTION The first decennium of the 21st century has been marked by environmental concerns gaining more and more attention. The acknowledgment of climate changes and natural resources' limits, the awareness of the negative impact of certain human activities on the environment and the necessity to reduce either this impact or these activities stimulated the interest towards natural and eco-friendly products. Over the last few years, the world wine market reveals the growing tendency to natural - organic or bio - wines'. As with many organic products, natural wine is becoming more widely available and offering more choice to consumers. Marketing sustainability and promoting environmental credentials becomes even a driving power in world wine (Brugarolas, 2005). New Zealand cultivates its "green" image (Hughey et al.,2005), Australia moves rapidly towards organic viticulture (Lockshin et al., 2008), Chile passes a country-wide sustainable code of conduct (Gibb, 2009).

In the meantime France overall did relatively little

environmental claims though during the recent years media generates a greater consumer awareness about natural wines, and the conversion process of vines into natural vines starts speeding in several regions. On the consumer side, several countries on the world wine market distinguish from others by an increasing percentage of natural wines in their total wine import; Japan is among these countries (Agence BIO, 2008). For French wine export, Japan holds the 6th position in volume and the 4th on value. France is the first wine supplier country for Japan (CIVB, 2009). Bordeaux wines are going through crisis and are willing to reconsider their strategy. The question we pose could natural wines be a solution for a small Bordeaux wine producer? If so, what are the conditions and constraints of this process? Are the existing supply chain practices advantageous for Bordeaux natural wines' export to Japanese customer? There exits numerous studies on supply chain related to organic products (Wycherley, 2002; Smit et al., 2008) or sustainable issues of SC practices (Verhagen, 2004; Carbone and Moatti, 2008), though not many on wines' SC. Most of the recent studies of supply chain practices in wine sector focused on the structure of the entire logistics chain and assessing the actors' performance (e.g. Chandes et Estampe, 2003); or explored the complex relationships that exist within the supply chain and the nature of stakeholders perceptions and (Monday, 2010). Little is known about specific characteristics of natural wines' supply chain management though "natural" wines by definition are more dependent on partners.

2


This paper proposes to partially fill this gap by presenting the different barriers and driving forces for organic wines supply chain practices focusing more on Japanese market, one of the world leaders in natural wines' import. In total, Bordeaux wines represent 25% in volume and 30% in value of total French wines import to Japan (CIVB, 2009). However, the Bordeaux natural wines have been facing challenges on the Japanese market (Bouzdine-Chameeva and Ninomiya, 2009), partly due to the supply chain problems. Our work focuses on understanding the role and interactions of major actors on Bordeaux wine scene - wine producers, wine merchants, wine brokers, wine syndicates – and major actors of wine distribution in Japan. On one hand we explore core and distinctive competencies of Bordeaux wine chain sector, to analyze marketing and distribution solutions of Bordeaux natural wines' producers; while on the other hand we attempt to tackle requirements to natural wines' supply chain on the side of Japanese importers. Our study has a double purpose : to model the natural wines supply chain model for Japan and identify the barriers and driving forces of the natural wine companies in the Bordeaux region, their supply chain challenges in Japan. 1 : DEFINITION OF "NATURAL WINES" There exist several different terms that are used towards eco-friendly wines: bio wines, organic wines, biodynamic wines, natural wines, authentic wines and even green wines (Krzywoszynska, 2009). Although we have chosen to use the term "natural wine" in our paper, we are aware of confusions with this term1 . We use this term for organic, bio or biodynamic wines without making difference. Below we briefly explain these differences though we consider that as the sector of natural wines is still small it is not important for our research to distinguish these wines for the analysis of supply chain issues. "Bio" wines are certainly a wrong term to use as only grapes are "bio" grapes. Nevertheless this term is frequently used for natural wines emphasizing the link with bio- products. "Organic" wines are produced by specific management practices that take care of the environment and soil. Synthetic chemicals and artificial fertilizer including pesticides are not permitted other than those specifically listed by the specific EU regulation of 2002. Totally organic wine does not exist, in either Europe or France. In fact, the term ‘organic wine’ is used to describe wines made from organically grown grapes (AB certificate in France), although other ingredients are not certified. Organic wine not only uses organic grapes but is

3


usually processed using the minimum of chemical intervention during the production process. There is, however, no agreed standard for this2. "Biodynamic" approach in winemaking is based on the writings of Rudolf Steiner, an Austrian philosopher who lived from 1861 to 1925. The underlying beliefs are based on the cycles of the moon and the alignment of the planets which dictate the days and the times of day at which certain activities are best carried out in relation to the planting, cultivating and harvesting of crops. In addition to adhering to organic farming principles with no use of pesticides whatsoever, biodynamic calls for the application of specific plant preparations created from animal manure. Nicolas Jolly explained in his book "Le vin, la vigne et la biodynamie" (1997) how the principles of Rudolf Steiner could be adapted to winegrowing to enhance the manner in which their wines expressed their terroir. While no scientific studies exist that could either explain the underlying science of bio dynamism or validate its ability to improve the quality of agricultural production beyond that of organic farming, over 600 winemakers throughout the world have adopted it (Carpenter, 2009). Several certification processes are put in place in France for organic and bio-dynamic wines as Biodyvin, Ecocert, Demeter. International wine competitions are organised specially for organic and biodynamic wines, for wines made from certified organic grapes, and wines made from transitional grapes (in the conversion process). When biodynamic wines were put to the test in a blind tasting by the major US financial magazine, Fortune, it was the biodynamic wine that was preferred by wine critics in nine of the ten pairings. 2 : THE NATURAL WINES SECTOR IN FRANCE The French National Agency BIO for the Development and Promotion of Biological Agriculture report (2008) indicates that in twelve years, from 1995 to 2007, the surface under bio vineyards in France has multiplied by 4,6 moving from 4 854 hectares to 22 510 ha. Moreover in the last five years the conversion of vineyards to organic or natural vineyards accelerates all over France, and the annual rate of conversion varies between 20-25% per year since 2006 (the conversion period takes three years). It becomes even more striking as the total surface of national vineyards in France reduces during the last years.

1

In a pure direct sense, the term "natural wine" - wine made by nature without any human intervention, will actually define vinegar. We use this term in a broad sense as a wine where human's intervention is limited and is in harmony with nature (it might be bio wine, organic wine, biodynamic wine and definitely green wine). 2

"No Organic wine processing rules in near future in the EU", Press Release of t International Federation of Organic Agriculture Movements – EU Regional Group of June 17th , 2010. Accessed on June 22nd, 2010 at http://www.ifoam.org/about_ifoam/around_world/eu_group-new/media/2010/PR_IFOAMEU_organic_wine_17.06.2010.pdf

4


Nevertheless organic grapes represent still only 3.3% of all French vineyards. The three major regions in France which reveal particular dynamism in this process are Mediterranean regions - Languedoc-Roussillon and Provence-Alpes-Côte d’Azur followed by Bordeaux (more precisely by Aquitaine region which involves Bordeaux area- see Figure 1.). Being on the honorable third place is pleasing though the reality of the process remains quite discouraging: the number of natural wine producers in Bordeaux is less than 300 (compared to total of 9000 winemakers in the region); the surface under organic wine farming is around 2000 ha (compared to the total of 118 900 ha) and bio certified wines are only 980 ha.

Figure 1. The surfaces under the bio wine production in France in 2007 (Agence BIO, 2008) The number of wine producers engaged in the conversion process grows though relatively quick growth of "natural" wineries is not always accompanied by a steady growth of quality despite a heavy certification process which is put in place. Why the process is so painful? 3 : BORDEAUX WINE CHAIN: SPECIFIC CHARACTERISTICS Bordeaux is the largest French AOC1 winegrowing area and the third largest wine-producing region in France. Traditionally the Bordeaux wine chain in is fragmented, includes various actors each having its own constraints and strategies (see Figure 2). Over the past 20 years, the number of winegrowers and wine producers in the area has decreased by more than half, falling from a total of slightly more than 20 000 in 1987 to almost 9 100 in 2008. 1

AOC – Appellation d'Origine Contrôlée. The AOC laws specify and delimit the geography from which a particular wine originates and methods by which it may be made. The regulations are administered by the Institut National des Appellations d' Origine (INAO).

5


Concurrently, the average size of estates has significantly increased, progressing from 5ha in 1987 to more than 13 ha in 2008, but almost 15 ha just for AOC wine producers.

Figure 2. Principle actors of Bordeaux wine chain1. During the past few years, the proportion of small estates covering less than 2ha (22% of the total and 1% of the area under vine in 2008) has considerably decreased to the advantage of larger domains, those of 20 ha and more, which now concentrate 23% of the total and 64% of the area under vine planted in AOC. Wine estates producing natural wines are normally of a relatively small size, these "natural" wineries rarely exceed 30 ha. So far the process of horizontal integration might not be a positive sign for the development of natural wines. Organic wine farming and biodynamic winemaking by definition could not be industrialized; it is easier to launch the conversion in small vineyards than of a large wine estate. 42 wine cooperatives and 6 cooperative unions regroup 43 % of winegrowers, in total they take 22% of areas and 23% of total harvests. Cooperatives exist on the principle of the mutualisation of production costs to preserve the small producers by bringing them a financial, administrative, technical and financial support. They also focus on improving the quality of wines by favoring the technical training of the winegrowers. Their place becomes particularly important for natural wine producers (Chabin 2008). CIVAM BIO 33 since 1991, or a more recent and small Bordeaux Bio Vignerons of Aquitaine unions are the vivid examples of heterogeneous reality of Bordeaux wine scene (e.g. Coop de France, 2007) . Around 100 wine brokers, more that 300 wine trade firms or wine merchants (CIVB Source, 2007) sell wines to customers.

Figure 3. Breakdown by type of dispatch during 2007-2008. 1

We frequently refer to wine growers and wine producers as winemakers. HSM is used as an abbreviation for hyper- and supermarkets.

6


On average, bulk purchases withdraw as such or bottled at the estate by the wine merchant represent 53% of dispatches from cellars (see Figure 3). Wine merchants' position in the Bordeaux wine chain is traditionally crucial as they represent for century an important economic and financial strength. They treat more than 70 % of the total wine commerce and export Bordeaux wines in more than 160 countries (CIVB report, 2008). Wine brokers take care of the En primeurs1 market transactions; their major role is in sales activities for chosen vineyards and also commercializing wines under brand names. Being at the end of the chain close to final consumers wine merchants ascertain consumers' preferences more than a wine producer. Wine producers are separated in the chain from consumers and thus they lack knowledge of the consumer market. The presence of an intermediate actor in Bordeaux wine chain confronts clients' desire, natural wines' clients prefers to know all the details of the wine production process and often meet personally a winegrower. Natural wines recently appear in wine merchants' offers, a small agency has been created recently to deal only with natural wines (Redsap - http://www.redsap.fr/). However this middle position of wine merchants is not easy either as they are squeezed on the one hand by the major retailers (hyper- and super-markets, growing hard discounts during the recent five years) and on the other hand by the fact that they lack the financial resources to manage effectively customer relationships (Chandes and Estampe, 2003). Moreover the competition with other retail channels as e-commerce which is rapidly growing during the last three years starts having an effect on wine merchants' position. The Internet wine market develops rapidly with growth rates of more that 30 % since 2007 according to the e-performance Barometer - sale of on-line wine (Bressolles, 2009). In France this market represents approximately 237 million euro in 2009; it is based on some 285 principal actors with the annual rate of renewal of about 7 %. "Pure players" specialized on the on-line wine commerce represent 32 % of Internet sales (several of them are located physically in Bordeaux area). The Internet sales of Bordeaux natural wines are tiny; they contribute less than 3% to total sales. 90% of buyers first visit the place and then purchase on the Internet. Nevertheless, this channel is growing mainly due to the information on bio winegrowers on the Internet which stimulates interest towards their wines and a visit the place with a potential purchase.

1

En primeur or in French – "marche des Primeurs" - is a method of purchasing Bordeaux wines early while a vintage is still in a barrel, before the wine is bottled - a year or 18 months prior to the official release of a vintage.

7


Bordeaux wine scene is quite complex; the relationships among actors have been cultivated through centuries and keep strong signs of old historical traditions. The question of wine chain integration is one of the major problems for Bordeaux wine scene. Bordeaux wine sector is on a collision course with the new changes in markets, new trends of consumption and new logistic channels as well (as hard discount, e-commerce, and Internet). Modernity challenges traditional prestigious Bordeaux core competency in wine. 4 : RESEARCH METHODOLOGY To analyze the main reasons of the slow rate of the conversion into organic vineyards in Bordeaux, and grasp the difficulties which wine growers and wine makers incur we adopted the approach of an interpretive case studies in eleven wineries in the Bordeaux region, with five wine merchants, and with a wine broker's agency and in parallel with three natural wines' distributors in Japan. The interpretive case study methodology (e.g. Eisenhardt, 1989; Pettigrew et al. 2001) refers more to ethnographic methods and focuses on investigating phenomena in its real –life context. Applying this methodology to atypical extreme cases in complex and uncertain conditions has given rise to several significant studies in the last years aimed at generalized findings (e.g. Flyvbjerg, 2006; Yin 2009). It has been shown that these cases could reveal more information than numerous representative cases clarifying causes and their consequences and leading to generalizing. To succeed with this objective we employ content analysis and critical impact technique (e.g. Porter 1980; Eden and Ackermann, 1998) as our research methodology. Through analysing a company’s comparative advantages, and difficult periods in its past, critical impact technique allows managers to better identify the internal constraints, strengths and weaknesses of the company. The links between cause and effect in this analysis enable managers to pinpoint competencies a methodology widely used in analysing critical company situations. Research design of the procedure starts with studying the questions used to enquire about the situation of the company, export policy and supply chain practices put in place. We proceed to interview managers and/or owners in a semi-structured format. Our aim is to explore with managers the interactions among the actors of the supply chain and to reveal any distinctive features of their competencies in this chain. Each interview lasted one hour and a half at average. 4.1: Bordeaux natural wines' chain (examples from the case studies)

8


The passionate wine producer 1 is the pioneer of biodynamic approach in Bordeaux region. The property of 10 ha with an annual production of 20 000 bottles is one of the first in Bordeaux; they celebrate 20 years of their biodynamic wine with the 2010 harvest. Their distribution channels are traditional for Bordeaux – wine merchants and retailers (cavistes only); they start selling via Internet using some pure players of the wine e-commerce sector. The wine is sold far beyond French frontiers - to Germany Austria, Belgium, UK, Italy Switzerland and Japan; The bio dynamic approach is employed in the big property of 80 ha of the wine producer 2 who is enthusiastic about the nature – oriented wine growing; the approach was put in place several years ago though grapes are not yet certified as organic one. In 2007 year, reputed by a very humid and cold summer, the property was on the third year of the conversion process. Diseases started spreading quickly on vines, and the owner of the wine estate (who is not a wine maker) made a decision to treat vines chemically not to loose the harvest totally; the conversion was stopped and the certification was not obtained. This example demonstrates the fragility of the conversion process; multiple risks Bordeaux winemakers incur while converting into organic wine-making. Bordeaux oceanic climate is not stable and lacks consistency compared to the Mediterranean regions of France, for example where organic farming gains more and more areas. Risks involved into the interdiction of chemical treatment of grape diseases in organic wine farming are quite high. Distribution channels for this winery are exclusively wine merchants and wine brokers (via the en Primeurs' market) as this property belongs to the old traditional Grand Crus chateaux in Bordeaux. They start selling via Internet channel however. Wine producer 3 possesses a small vineyard of 10ha in Sauternes area of Bordeaux famous sweet golden wines. The annual production of this biodynamic property which has one fulltime staff member who is an owner /manager is around 160 barrels. Wines are sold via bio agencies, wine tasting and following "a word of mouth". No special efforts of developing distribution channels are made. Wine producer 4 of the Bordeaux Grand Cru wine estate of 14,5 ha pursues bio dynamic approach since 1997 with a consultant who visits the winery several times par week.. However they possess no official bio-dynamic or organic label - "those who know our wine knows and I am not keen on making publicity of my environmental concerns", stresses the owner. In a long run there were several critical incidents which ensured the wine producer in this choice. In 2002 and 2004, the weather was so bad that without treatment, everything 9


would be lost, and they sprayed vines strict minimum. In the promotion of their wines a strong accent is put on the harmonious approach to winemaking nurtured in the property for many years. Distribution channels of are mainly wine brokers and wine merchants, however direct sales (to Japan in particular) and e-commerce are also considered as the channels the company widely uses (see Table 1). All wine producers emphasize the positive role of wine merchants if they distribute via this channel; wine merchants are praised for their deep knowledge of market and distinctive competences in distribution, particularly in foreign wine export. None of the wine producers we interviewed uses HSM retailers; Small-size natural wineries with a lower output compared to the conventional wine production yield, are not able to guarantee the quantities required by the HSM. Cavistes, restaurants, bio agencies, e-commerce better meet their distribution needs.

Table 1. Examples of SC practices used by Bordeaux natural wines producers. Moving forward along the Bordeaux wine chain in our study, we interviewed two wine merchants. Both of them called our attention to relatively low figures of Bordeaux natural wine production. They diversify their portfolio adding natural wines offers, though they are not eager to make special efforts for the promotion of this particular part of the market. Wine broker, whom we met next, confirmed the position of wine merchants - natural wines are still almost absent at "en Primeur" market. On one hand it is explained by heterogeneous quality of Bordeaux natural wines and on the other hand, by insufficient quantity compared to the average transaction volume. The small wineries can’t grow up rapidly; they first seek the market to sell, invest afterward and only then increase the production. Wine broker's objectives are not entirely compatible with natural wines offers. On the contrary to the wine merchants and wine brokers, the unions of natural wine producers are ready to make a great effort into developing new distribution channels and invest into the maturing customer relationships. However there is no comparison in the financial situation of 10


wine merchants, brokers and natural wines' unions, the latter have limited budget and stay within rigorous financial constraints (Cholette, 2010). Our findings show the strong heterogeneity of the Bordeaux natural wines' production sector, the complexity of relationships among the wine chain actors of and considerable differences in solutions found by wine producers dependent on the core competencies, traditions and attitude towards innovation. 4.2: Japanese natural wines market In 1969, French government (Ministry of agriculture and fisheries) set up their liaison office in French embassy in Tokyo. At that time, wine was drunk at the hotels or the French luxury restaurants only at special occasions such as wedding reception. The liaison office became independent as SOPEXA Japan in 1985 with increasing the export of French food and wine. However, it could be said that there was no wine market in Japan until 1990’s, Sopexa was supposed to enlighten the Japanese people. They started initially with wine brochures, booklets, with organising seminars; then in order to expand more the wine market in Japan, SOPEXA Japan tried to establish the professional vocation of a sommelier. Sommeliers are close to the consumers, the importance of their role in educating consumers became evident to SOPEXA They invited French top sommeliers and chefs to teach French wine and cuisines in order to prevail wine, started to hold World’s Best Sommelier Competition twice a year in Paris. In 1995, a Japanese sommelier, by the name Mr. TASAKI Shinya, gained the world sommeliers' championship. Media promoted this success all over Japan, and that was a crucial point of growing interest towards wine among Japanese. Sommelier had been firstly recognized as a profession founded upon specialised training and as specialist in Japan1. In 1998, a red wine boom, ‘French paradox’, occurred in Japan, the consumption of wine had increased rapidly. The sales of wine reached 298 thousand kl (see Figure 4), 242 thousand kl of bottled wine was imported, 42.6 % of imported bottled wine was from France (Nikkan Keizai, 2000). In the end of 90’s, many Japanese went to Bordeaux and Bourgogne to study wine production and tasting, received a classical wine education, they went back to Japan, became sommeliers, wine importers, cavistes, opened wine bars2. A few Japanese importers noticed the existence of natural wine. Although they were generally surprised at the different

1

These days, the term " sommelier", became a cliché in Japan, there exist a green tea sommelier, a vegetable sommelier, a food sommelier etc. Private associations find it prestigious to have sommeliers. 2 Japan holds the largest number of sommeliers in the world; 13,409 sommeliers, and 10,435 wine advisers, who sell or import wine (Japan Sommelier Association website ( http://www.sommelier.jp ) accessed in the March 1,2010).

11


taste from the one they had had before (Arai, 2005; Ohashi, 2004), caught by the fruity flavour and charm and started to import it little by little.

Figure 4. Wine consumption trends in Japan. (Source: National Tax Agency Japan) The first signs of "Natural Wines’ boom" in Japan appeared around 2005. The term SIZENHA Wine’ is used in Japan instead of just ‘organic’ or ‘nature’ wines. This word can be translated as natural wine or nature-oriented wine, because organic or natural wine might be used formally and commercially. A monthly magazine specialised in natural wine, called ‘Real Wine Guide’ was first published (Hori, 2005), another popular magazine put together a special issue on Bio and Natural Wines. In the last three years many cavistes decided to specialize in natural wine. The importers shift over into not caring much about the AOC, or the bio certificate; they rely on different sources of information, such as cavistes, bistro in Paris, BIO salons, personal network, etc.; they definitely favour direct transparent long-term relationships with natural wine producers. The key for the importers is the quality of wine. They consider that to make good wine winegrowers should cultivate good grapes in an organic way: only a wine producer close to the nature, passionate about the vines is capable to produce authentic natural wines. 4.3: Supply chain of Japanese importers (examples from the case studies) Racine (Tokyo) is one of the most famous natural wine importers in Japan. Their sales in 2006 are about 600 Mln JPY (6 Mln USD) dealing with 150 wineries in France, Italy, and Slovenia. Natural wine" is said to be very sensitive to the temperature and humidity. Mme. Goda, the owner of Racine, the holder of DUAD from Bordeaux, is the pioneer in the new approach of total quality temperature control - from its original storage place to the consumer in Japan. She checks the warehouse with thermometer and hygrometer, and controls the wineries with whom she deals that only reefer containers are used for wine transportation. 12


The transportation fee of one dry container is 130,000 JPY (1,300 USD), on the other hand, reefer container cost 800,000 JPY (8,000 USD). They sell to hotels, restaurants, and retailers. The importers trade exclusively with those who provide the mandatory conditions and meet the strict requirements. They transfer the wine a reefer truck from May to September in Southern area. The cavistes keep the wine in air-conditioning room. The export of natural wine is extremely costly. Therefore the retail price varies from 2,000 JPY (20 USD) to 4,500 JPY (45 USD). They pay 50% to the winery after the harvest and 50% after the bottling. Importers visit the vineyard twice a year at least. Maintaining close relationships with suppliers is one of the core competences of Racine. Another example of an important actor of the Japanese wine supply chain is J.F. Hillebrand (http://www2.jfhillebrand.com/ ) who built up the global logistics network of wine in this decade. The importers just make a phone call to wine producers and to JFH. JFH makes arrangements on everything under EX-Works conditions from every region in wine countries to Japan in reefer containers, from one box to several cases. The importers can buy from every region and therefore concentrate themselves only on trades. Small wineries in France with whom Japanese importers deal, vary in size: from 3 ha run by the one-man farmer to more than 20 ha. Respectively the volumes exported by French wine estates to Japan are quite different. Sustainability of relations in the supply chain is both, the objective and an aspiration, of Japanese wine importers. 5 : BARRIERS AND DRIVING FORCES OF WINE SUPPLY CHAIN PRACTICES The wine sector has come to the idea of the importance of Supply Chain Management to a branch's overall performance. Wine producers become more aware of the link that centralising a significant number of value-added operations, they can move to a supply chain integration and unification approach and finally lower their expenses on logistics (Chandes and Estampe, 2003). Two out of the four models suggested in this recent study on the logistics issues on wine supply chain performance describe well the case of Bordeaux natural wine exports (see Figures 5 and 6). Some "natural" wineries once sold their grapes or must to the wine merchants or the cooperatives at the low price. They made a decision to bottle their wine by themselves, at the same time; they had to sell by themselves, bearing the risk to sell. Small ‘natural’ wineries in France exporting abroad, entrust the one nation’s trade to one importer. As a general rule, wine export from one country to Japan is realized by only one importer. This practically gives the exporters the exclusive right to sell. However, in the case of small wineries it is frequently a verbal promise based on a human relationship and trust. 13


Figure 5.Grand Crus supply chain model ( adapted from Chandes and Estampe, 2003).

Figure 6. Wines marketed directly by the wine producer (adapted from Chandes and Estampe, 2003) To succeed in the Japanese market, the small ‘natural’ wineries need to build deep trustful human relationship with the Japanese importer; sharing the same values and philosophy. Quality and quantity of wine might vary from year to year, this is particularly true for natural wines. With close links in supply chain, the importers understand what happens in the vineyard they deal with. If there is no close relationship between a winery and an importer, the Japanese importer might switch to another supplier (Ohashi, 2004). Wine industry in Bordeaux region has developed, being based on merchant-indirect supply chain system. Merchants control the quality, stock, physical distribution (logistics) abroad, sales promotion; they have their own brands and bear the risk in the whole selling processes. Under these circumstances the producers do not have to care about selling, can concentrate on producing. The importers in Japan can buy the Bordeaux wine easily and feel relieved, 14


leaving everything to merchants. Formerly, it is difficult to access the small regions because the importers need setting up the SC system by themselves to maximize a container load.

Figure 7. The comparison of the Japanese supply chain for Bordeaux and other than Bordeaux wine regions in France. On the other hands, natural wine requires a different type of supply chain system, which must be ‘face -to –face’ direct network. Natural wine making need more labour, the crop is unstable depending on the weather, threatened by insects causing damage to crops, the tastes vary every year. The producers need to let consumers understand their hardship and philosophy (Olsen et al., 2006). In Japan natural wine importers play an important role in transmitting this information to the consumer. Natural wine importers bring producers to Japan, hold tasting events in Tokyo, in small cities, providing an opportunity to meet with and talk to the consumer. The consumers became their fans, and that makes the supply chain functioning. Several companies of wine management and wine consultancy have been established in Japan to facilitate interactions along the supply chain of natural wines – Oeno Connexion and BMO created in 1997 and 1998 respectively are the examples of this trend (see Figure 8)

15


Figure 8. An example of the information/product flow chart of Japanese SC actors. Bordeaux natural wines are facing also marketing difficulties in Japan as a wine producer is separated from a final consumer. The image of Bordeaux wines on the Japanese market which is directly associated with luxury, complexity, magnificence is in discord with the image of simplicity for natural wines. Educational efforts become essential, direct links between wine producers, importers and consumers appear to be crucial to change this situation and adapt the image of Bordeaux natural wines nurtured in long-standing nature-oriented traditions of terroir. A thorough, clear understanding of the client on the end of the distribution chain could become a core strength for small wineries; a conceptualisation of decision-making based on learning about the importer's side would facilitate and improve the distribution channels. This would require a closer collaboration along the supply chain. CONCLUSIONS AND PROSPECTS FOR FUTURE STUDIES In the research we have presented the analysis of Bordeaux natural wines supply chain focusing on the Japanese market basing on several case studies performed in Bordeaux region and in Japan. A double purpose of our study was to identify the major characteristics of natural wine companies in the Bordeaux region, their current distribution channels, and on the other hand, to understand the functioning of the Japanese supply chain and Japanese importers behaviour. The choice of the Japanese natural wines market is explained by the enthusiasm and the openness of this market to the eco-friendly products and the strong presence of Bordeaux wines on this market. During the last years Bordeaux wine sector is on a collision course and needs to revise its strategy in the world markets - natural wines might be seen as a solution to work up a new axe for future growth. Sustainable quality issue is seen as a driving force on the side of wine producer and on the side of the Japanese market. Earlier it has been shown that wine quality appears to be the key factor (and therefore, a core competence) in the stable success of a winery. In fact, all decisive themes focus on reinforcing the question of managing stable wine quality as a key to attain stability of the winery development (Bouzdine-Chameeva, 2006). In noting the varied and sometimes complex relationships within the supply chain, the research has confirmed that maintaining high quality remains the driving force for Bordeaux natural wine producers. Nevertheless our findings confirm that there are several barriers of for Bordeaux wines to enlarge their presence on the Japanese market. First, instability of Bordeaux oceanic climate and lack of consistency compared to Mediterranean regions of the south of France increases

16


risks for producers in the conversion process. Risks involved into the interdiction of chemical treatment of grape diseases in organic wine farming are quite high. Secondly, the multifaceted and fragmented supply chain system of Bordeaux wine scene with wine producers, wine merchants, wine brokers, retailers, wine associations, union and syndicates,

growing e-commerce solutions creates barriers to successful evolution of

tradition SC practices; its complexity puts forward the issue of supply chain integrity. Bordeaux natural wine producers continue to produce and promote wines more or less as did their grandfathers, with a strong emphasis on "terroir" and on the particular vines that bring out the true qualities of this "terroir", with the use of traditional intermediate actors in the supply chain. However, this approach reveals to be in some sense indolent, as no longer corresponds the widely varying reality of international demand. As a result, Bordeaux natural wine producers have to urgently reassess their traditional SC practices and image. A better understanding of “changing customer needs” could better help Bordeaux natural wine companies reformulate their individual and collective goals, and redefine their individual and collective strategies and improve their position. Direct links between producers and importers/consumers in Japan are seen as a strong driving force of the SC development for Bordeaux natural wines in Japan. The findings confirm that successful penetration into the Japanese market bear a strong relation to the active construction of distinctive SC links for a winery. A thorough, clear understanding of the client on the end of the supply chain becomes core strength for small wineries and a conceptualization of decision-making based on learning about the structure of the supply chain management on the importer's side will facilitate and improve the supply chain solutions. A more profound analysis of the boundaries and organizational forms appears necessary. Future work will examine what forms the direct links might take in the supply chain, how they can be defined, developed and implemented in the complex situation of a Bordeaux natural wine chain. REFERENCES. o

Agence BIO – Agence Française pour le Développement et la Promotion de l’Agriculture Biologique (2008) La tendance est bio ! (Statistics on the bio wines). Accessed March, 2010. From http://www.agencebio.org

o

Arai J. (2005), Budo batake de nagagutsu wo haite (Wearing long boots in wineyard), Tokyo: Shueisha International. 17


o

Bordeaux Wine economic profile (2008), Director of the report R. FEREDJ, Conseil Interprofessionnel du Vin de Bordeaux (CIVB) "Economy & Studies"department, 10p.

o

Bouzdine-Chameeva T. (2006) ”Modelling of distinctive competencies in strategic development of Bordeaux wine sector: comparison based on causal representations”, British Food Journal, Emeralds, Special issue on Wine strategies, Vol. 108, n° 4, pp. 273-289.

o

Bouzdine –Chameeva T., Ninomiya M. (2009), Challenges of Bordeaux natural wines, Proceedings of the 4th Interdisciplinary and International Wine Conference, 'Bacchus goes Green', Dijon, July 8-10.

o

Bressolles G. (2009), e-Performance Barometer, online wine sales, BEM, 39p, www.ePerformance-Barometer.bem.edu .

o

Brugarolas M., Martinez‐Carrasco L., Martinez Poveda A., and Rico M. (2005), Determination of the surplus that consumers are willing to pay for an organic wine”, Spanish Journal of Agricultural Research, Vol.3, n°1, pp. 43–51.

o

Carbone V. and Moatti V. (2008), Greening the supply chain: preliminary results of a global survey, Supply Chain Forum: An International Journal, Vol. 9, n° 2, pp. 37-49

o

Carpenter M. (2009), Château Larrue: the growth of a biodynamic wine brand, Case Study, ECCH, 11p.

o

Chabin Y. (2008), Sustainable Development and Wine Cooperatives: stakes and perspectives, Colloque de la Société Française d’Economie Rurale, Les entreprises coopératives agricoles, mutations et perspectives, Paris.

o

Chandes J., Estampe D. (2003), Logistics Performance of Actors in the Wine Supply Chain, Supply Chain Forum: An International Journal, 2003, Vol. 4, n° 1, pp.12-27.

o

Cholette S., (2010), Postponed Practices in tehe wine industry: adoption and attitudes of California wineries, Supply Chain Forum: an International Journal Vol. 11, n°1

o

CIVB (2009), 80 destinations pour les vins de Bordeaux: Analyse des exportations de 1982 à 2008, The 8th edition, n° 277. Conseil Interprofessionnel du Vin de Bordeaux.

18


o

Coop de France (2007), Coopératives : comment s’engager dans le développement durable ? , Coop. Infos, n°16, pp. 3-16.

o

Delmas M.A, Grant L. (2008), Eco-labeling strategies: The eco-premium puzzle in the wine industry, AAWE Working paper, n° 13, 33p. Accessed March 30, 2010 at http://www.wine-economics.org/workingpapers/AAWE_WP13.pdf

o

Eden C., Ackermann F. (1998), Making Strategy: the journey of strategic management, London: Sage.

o

Eisenhardt, K. M. (1989). Building theories from case study research. The Academy of Management Review, Vol.14, n° 4, pp. 532-550.

o

Flyvbjerg, B. (2006). Five Misunderstandings About Case-Study Research, in Qualitative Inquiry, Vol.12, n°2, pp. 219-245.

o

Gibb R. (2009), Chile unveils sustainable code, Decanter, 10 septembre 2009.

o

Hori A. (2006), Rialu wain gaido Burugounyu (Real wine guide of Bourgogne), Tokyo: Shueisha International.

o

Hughey K.F.D., Tait S. V., O'Connell M. J. (2005), Qualitative evaluation of three ‘environmental management systems’ in the New Zealand wine industry", Journal of Cleaner Production, Vol. 13, n° 12, pp. 1175-1187.

o

Krzywoszynska A. (2009), Green winemakers : the role of nature in the ethics of Italian organic winemaking. Proceedings of the 4th Interdisciplinary and International Wine Conference, 'Bacchus goes Green', Dijon, July 8-10.

o

Lockshin L., Remaud H., Mueller S., Cvyl P., (2008), “Do Australian wine consumers value organic wine?”, 4th International Conference of the Academy of Wine Business Research, Siena, July 17‐19.

o

Monday A., Wood-Harper T. (2010), Exploring the supply chain of small and medium-sized South Australian wine producers, Supply Chain Forum: an International Journal, Vol 11, n°1 (in press).

19


o

National Tax Agency Japan (2008), Sake no shiori (Statistics on the alcoholic beverages).

Accessed

March,

2010.

From

http://www.nta.go.jp/

shiraberu/senmonjoho/sake/shiori-gaikyo/mokuji.htm o

Nikkan Keizai (2000), Tsushin Co.,Ltd. Edition, p.

o

Ohashi K. (2004), Shizen-ha wain (Natural Wine), Tokyo: Shibatashoten.

o

Olsen J., Nowak L. and Thach E. (2006), Integrating Environmentally Friendly Behavior with Hedonic Consumption: The Case of Organic Wine, 13th Academy of Marketing Science World Marketing Congress, Verona, Italy.

o

Pettigrew A., Woodman R. W., and Cameron K.S. (2001) Studying organizational change and development: Challenges for future research, Academy of Management Journal, Vol.44, pp. 697-713.

o

Porter M. E. (1985) Competitive Advantage, Free Press, New York.

o

SOPEXA JAPAN (2010), http://franceshoku.com, accessed in the 30 March 2010.

o

Smit A. A. H, Driessen P. P. J. and Glasbergen P. (2008) Constraints on the Conversion to Sustainable Production: the Case of the Dutch Potato Chain Business Strategy and the Environment, Vol. 17, pp. 369–381

o

Storstad O. and Bjorkhaug H. (2003), “Foundations of production and consumption of organic food in Norway : common attitudes among farmers and consumers?” Agriculture and Human Values, Vol.20, pp. 151‐163

o

Verhagen H., (2004) International Sustainable Chain Management: Lessons from the Netherlands, Benin, Bhutan and Costa Rica, Netherlands International Partnership for Sustainability [NIPS] and the Royal Tropical Institute; Amsterdam: KIT Publishers.

o

Wycherley I. (2002), Managing Relationships in the UK Organic Food Sector, Journal of Marketing Management, Vol. 18 n° 7/8, pp. 673-692.

o

Yin, R.K. (2009). Case study research: Design and methods (4th ed.). Beverly Hills, CA: Sage Publishing.

20



Strategy and supply chain design for the Mexican automotive sector Miguel Alejandro García Vidales Facultad de Ingeniería, Universidad Autónoma de Querétaro ([email protected]) Rocío Montserrat Campos García Facultad de Ingeniería, Universidad Autónoma de Querétaro ([email protected]) Ovidio González Gómez Facultad de Ingeniería, Universidad Autónoma de Querétaro ([email protected]) Eduardo Betanzo Quezada Facultad de Ingeniería, Universidad Autónoma de Querétaro ([email protected])

Abstract

Mexico is a country where firms face several problems on industrial logistics, It has been reported that around 20% of total costs is derived to logistic costs in the auto part industry. The aim of this paper is to analyze the functions of logistics areas in the supply chain management of companies in the Mexican automotive sector. The outcome of this analysis might help design an optimal strategy for the supply chain management. The results of a random survey, show the needed data to design a new supply chain strategy, such as the administration of a logistics network, focused on the Mexican infrastructure and conditions, and this paper finally introduce a new integrated scheme to manage the logistics network design problem for the automotive industry.

Key words: supply chain, automotive sector, logistics, network design, Mexico. 1


1: INTRODUCTION It is well known that Supply Chain Management (SCM) is increasingly seen as a set of practices aimed at managing and coordinating the entire chain from raw material suppliers to final customers. Supply Chain (SC) integration is considered one of the most important factors to improve industrial performance. Over the past decades, one of the main issues in SCM literature has been the role of integration as a key factor in achieving improvements, (Vaart et al., 2008). The role of the automotive industry is not focused nowadays on producing more at the lowest cost, but, instead, on selling as fast as possible the largest amount of units at the lowest cost in accordance to demand. This, however, should be carried out in a production set up where the supplying companies are increasingly sending inputs from distant places. In spite of the correspondent increase in transportation costs this situation produces, the final goods still must be offered at the market sanctioned price, which poses a huge pressure on the requirements to reduce logistic costs at minimum. Moreover, Latin American automotive companies have become members of global supply chains due to the continuous and increasing globalization process. In fact, there is a great difference on logistic costs, depending on the position of the firm within the global production chain and the final destination of their products. Whether the firm is engaged in the extraction of raw materials; the production of intermediate goods like parts for assembly lines; or in the production of finished goods to retail markets, makes a difference on the logistics activities, control and costs assumed by each firm. Suppliers and assembly plants operate as part of a comprehensive system and, therefore, each firm is vulnerable to disruption by external causes as, for instance, the bankruptcy of clients or suppliers. This vulnerability has become a key variable in decisions relating to the selection of suppliers and has the potential to exclude specific companies and even to specific countries, thus hindering them of the potential economic gains from such trades. This vulnerability represents a potential cost that can easily wipe out any advantage that companies have on prices and or location. Supply chains are complex dynamical systems triggered by customer demands (Sarimveis et al., 2008). Moreover, the companies involved have a responsibility to provide quality products and ensure that the product will arrive in time and in good conditions after transportation and distribution stages. The activities of supply chains are placed as orders and are concluded when the customers receive the goods they have paid for. In that sense, a 2


supply chain system is a very complex network that spans from the final customers back to the retailers, wholesalers, distributors, manufacturers, suppliers, and ultimately to the raw material producers, (Ösbayrak et al., 2006). For a supply chain to work efficiently, proper links as well as close cooperation and coordination of these stages are required so as to enable all the elements of the supply chain network to be profitable. Nowadays, the concept of SCM has evolved from a narrow perspective, related only to material flows, to a broader view, encompassing material, information, financial and technical flows, both within each organization and among organizations (Caldeira et al, 2009). The SCM is a way of obtaining vertical integration benefits without its formal ownership costs and with a proper selection of equipment, machinery, buildings and transportation fleets. However, efficiency of supply chains mostly depends on management decisions, which are often based on intuition and experience (Sarimveis et al, 2008). Much of the research on logistics has been conducted on large international enterprises and it might be misleading to use its findings to improve SCM in small and medium companies. This research, therefore, focuses on understanding the context of the Mexican automotive suppliers (medium and big firms) in the specific field of logistics. The supply chain strategy can be carried out by replicating a formerly proven model strategy that has been developed by other firms. However most of the times, it is impossible to adapt external strategies, given the fact that these might be designed for specific and incompatible environments, needs or contexts. In recent years, the constant emphasis on productivity gains and customer satisfaction, has led to rapidly changing business environments characterized by time compressed supply chains, and by alliances and mergers acquisitions. In turn, these facts highlighted the importance of properly designing or redesigning the production and distribution networks of manufacturing firms. A growing emphasis on a new collaboration on line, technologically advanced manufacturing, just-in-time pick-ups and deliveries, is also amplifying the role of SCM as a strategic tool for competitiveness. As suppliers reveals the performance within the automotive production process, this paper addresses the problem of poorly designed supply chains or logistics networks for manufacturing firms operating in the State of Aguascalientes, Mexico. In that sense, we define a logistics network is a set of suppliers, manufacturing plants and warehouses organized to manage the procurement of inputs, their transformation into finished products and distribution to customers. Thus the aim of this paper is to analyze the functions in 3


logistics departments in a selected number of companies of the Mexican automotive sector, in order to contribute to the design of an optimal strategy for the supply chain management.

1.1: Background Although no national automotive industry has been developed in Mexico, national policies have been implemented along the last 50 years in order to attract foreign investment to this industry. The shifting national policies regarding the automotive industry, in general, have all allowed that this activity became one of the most dynamic and set Mexico among the main economies producing automobiles and auto-parts. The first national specific policy instrument (decreto) in this area, in the early sixties, pursued the strengthening of the automotive industry for the domestic market. As the economic national model shifted toward a free trade economy, the market was open for parts and vehicles since the middle eighties and was reinforced later on by the NAFTA (North America Free Trade Agreement) in 1994. This trend continued and is completed in 2004 by a total liberalization of trade in regards to this industry (Vicencio, 2007) Nowadays, the industry as a whole (parts and vehicles) stands for 2,5 % of GDP, 1,6 % of national formal employment and 19% of total Mexican exports (Vicencio, 2007). Total employment in the industry of parts amounts for 89 % of the automotive industry as a whole and 9 % of total Mexican exports. The economic activities with more foreign investment are concentrated in the production of parts for the automotive industry, 34% of the total direct foreign investment channeled towards this industry (Bancomext, 2010). There is no doubt that, both the terminal industry and that of its parts are two of the main economic activities in Mexico. It should be noticed the structure of the auto-parts industry. According to the national association of automotive part producers in Mexico (INA, for Industria Nacional de Autopartes), there are 600 firms of this type in the country. By size, they can be classified in 230 large, 162 medium and 208 small sized firms. Given the weight of the automotive industry in Mexican GDP, changes in its efficiency may lead to a faster growth which in turn may generate even more jobs and participate in the creation and distribution of wealth in the country, contributing the integration and development of the sector. Nevertheless, it should be mentioned the risk of a deeper Mexican foreign economic dependence. Surely, technology plays an important role in the market behavior of firms producing automotive parts in Mexico, within the global context. But most technological development 4


and innovation are produced outside this country. In contrast, soft technology, logistic included, is a field which can be controlled within the boundaries of the country. This paper aims to look into the intra and inter-firm relations covering the integration of productive SCM in the automotive industry as well as the practices that dominate the future of the parts industry and spares-automakers. We identify and evaluate the factors that companies exploit to obtain logistic advantages in one of the most dynamic cities for the Mexican automotive industry, Aguascalientes. To achieve this goal, we conducted an original survey, combined with the analysis of external factors such as services in the region, since they may promote logistics integration of internal functions within the firms, and coordination of external activities that are outsourced to logistics providers. We refer to this combination between internal activities and the coordination with transportation firms as the logistics performance of the industry by assessing it in the selected companies subject to this research. The factors that mostly affect the logistics of automotive parts and components are closely related to production systems and inventory control. We focused the research work on: processes and systems in the fields of information, production, inventory management, management of systems, and packaging. Several authors confirm that the transportation cost has lost significance when it is considered as a factor in industrial location (Hanjoul et al., 1983). Even though new technologies to do business (e-commerce) acquire an increasingly important role in the internationally traded automotive parts, they may not eliminate spatial barriers arising from the actual transfer of goods, because transportation costs will remain as a "parasite": it maintains the premise that transportation is not free, nor can it be done instantly. Hence, the logistical advantages relate only to the physical transportation of goods, and firms respond with a certain level of quality to the logistical requirements imposed by the logistics chain from producers or distributors within the possibilities derived from the quality of services and infrastructure. Moreover, the logistical advantages depend on the dynamics generated around industrial areas where high levels of expertise have been achieved for the automotive industry. Some regions in Mexico have already developed this kind of expertise, Aguascalientes amongst them. From this perspective, we assume that Weberian theories on industrial location have been largely overcome by considering the optimal location in terms of mainly transportation costs (Weber, 1928), and we extend the concept of externalities (positive and negative) or agglomeration effects to include technology, infrastructure conditions, and services that enable the industry to solve logistical problems. In a broader context, we take a holistic 5


approach that rests on the quality of the sector’s capital and networking movement. The function of these logistical advantages is to avoid any rupture in the continuity of flow quality (or even better, to take advantage of it), which is defined by traffic rules established in a chain. In this regard, we believe in the importance of a network which aims to maintain the quality of movement so tight, preventing the generation of some type of loss, goods damage, or costs of "lack of quality" of the processes. For the type of firms, such as those in the automotive industry, quality of services is a key factor in location, where the main interest is to increase the productivity of firm resources through a global reduction of the cost of goods movement, (Betanzo 1990). 1.2: Overview of the automotive industry. The European automotive industry is currently facing a threat for survival in the face of increasing global competition and a permanent over-supply in the automotive market. New vehicles must be launched onto the marketplace with a minimum of delay. It was estimated that an automotive manufacturer in Europe lost US$1.8 billion in profit (before gaining its market share) as it remained one year behind its competitors in introducing a new model to the market (May et al., 2001). However, a small number of firms involved are able to fulfill their logistics strategies for the supply of products. The automotive industry is one of the most dynamic sectors in North America, Europe and Asia. Mexico's auto industry presents a similar situation, for this industry behaves global. Therefore, every firm in the industry is in need of continuous technological innovation in order to maintain its share of the market. It should, thus, enhance the efficiency of vehicles design (in the case of vehicle producer firms), increase safety, reduce production costs and improve customer service after sale. This industry has also developed advanced technologies and brought new technological improvements in production systems and logistics management which seeks to cut the cycle that range from new vehicle design to production and distribution to sales agents. To meet increasingly competitive markets for multinational companies, new forms of production and organization have forced large global manufacturers to embrace new reengineering philosophies and to implement quality control. This has occurred by significantly restructuring the way of producing and transporting inputs, the way of storing parts or components, the procedures for staff training, and even developing new tools for telecomputer data transmission under the "zero paperwork" ideal (Betanzo 1990). Mexico is a country with several problems that affect industrial logistics such as lack of road infrastructure and poorly designed streets, avenues and highways, which result in damaged, 6


congested and most importantly, unsafe movements in freight transportation. In addition, the major rail logistics corridors (Kansas City and Ferromex) do not cover the entire Mexican territory, but only the commercial interaction between the northern border and the center. In the Latin American region, Mexico is one of the countries with the largest foreign direct investment in the automotive sector. Still, competition from other locations is always present and, in order to optimize costs and raise productivity, an increase in the level of efficiency for the logistic strategies design has become a must. The evolution of the automotive sector in each Latin American country is heavily influenced and supported by the industrial national policies. Their aim is to develop a locally integrated healthy industry, increasing the volume of vehicles annually produced, and able to compete under the standards of international quality.

2: METHOD

Most previous studies on SCM are based on anecdotal evidence or case studies (Wagner, 2006). Our study is based on the outcomes of an exploratory survey, consisting of personal interviews, to 12 companies in this sector. The selected companies had at least 400 employees and up to 6500. For their firm size, it can be presumed that they have significant interest in their SCM. New measures and a fully standardized survey instrument were developed in five sections. The first one is focused on logistics costs and it is designed to detect the most costly problems of the supply chain management; the second section is devoted to purchases and sales activities. The third section consists on production activities; the fourth one is aimed to the storage processes; and the fifth one to transportation and distribution. Gathered information allows us to identify the activities of the supply chains for surveyed firms of the automotive sector in Aguascalientes, Mexico. Bearing in mind the proposed search for a proper logistic strategy, improvements of the logistics network can be drawn from the understanding of the main current shortcomings and, from there, the elements for a formalized optimization model can be depicted in order to solve or ameliorate the problems of the supply chain that nowadays exists in the Mexican companies.

2.1 The companies

The twelve companies surveyed are involved in the production of automotive parts for the assembly industry. They all belong to the first tier (T1) group for Nissan, in Aguascalientes, 7


Mexico. They represent a group of companies with similar characteristics, expressed some involvement in growth strategies to increase production, and they all stated are working with JIT (just in time) procedures.

3: DISCUSSION AND RESULTS

Companies in the automotive sector in Mexico were asked about the logistics activities that they are performed, we analyzed the purchase and sales activities, production, storage and transportation and distribution, questions have five possible answers, where level 1 is the simplest form they can perform the activity and level 5 is the most advanced one (according to current technology). The five levels for each response were based on the tools described in the literature for each activity. For buying and selling activities, we found that 10 (83%) firms have a delivery schedule to customers, but it is not detailed. A non-detailed planning can cause problems with customers due to unpredicted failures in delivery times; the firm can lose the costumers trust or, at worst, may lose part if not all of the market. Another question addresses the information flow. It is very important in a supply chain because the harmony of the whole supply chain depends on it. Only 7 (58%) firms have updated data on order status to customers and it is incomplete (these do not have traceability on transport, or distribution). Moreover, only 9 (75%) firms report performing a proper documentation of the sales forecasts results but they do not share information with other areas of the company on the supply chain, this point shows that the majority of the firms needs to improve the information flow to avoid problems that might threaten their very existence in the market. The suppliers issue is one of the most important in logistics; if firms have the right supplier, it can almost be granted that they will always have their inputs in time and under the agreed standards of quality. In the study, delivery times from suppliers are monitored in only 6 (50%) firms. Only 6 (50%) firms reported shipping notices of orders in advance, in order to arrange personnel for receiving inputs: As a criterion for the selection of suppliers, in 10 (83.4%) firms the interviewed mentioned the use of global performance indicators for evaluating potential suppliers and the formal analysis of supplier’s performance factors that might influence suppliers future response.

8


Regarding the status of orders to suppliers, 6 (50%) firms have available data. Communication for data exchange with suppliers is conducted by email, fax and telephone in 6 (50%) firms. No advanced real time data interchange was reported. As SCM cannot be understood isolated from the production itself, but it should be considered rather as a constitutive part of it, we asked about the production strategies of the firms. Only 7 (58.3%) firms apply several improvement techniques in areas related narrowly to production. But even there, there is no continuous monitoring. 6 (50%) firms have a formal planning system for the different activities that are needed for the completion of each order; 9 (75%) know the projected production capacity and are aware of the actual utilization and the progress of the production program; 6 (50%) reported the use of a computerized management system where a central station automatically receive data from each production station and it can be immediately obtained the progress for each order. Proper production planning allows optimization of resources and ensures customer satisfaction by delivering orders with the required quality. In an ideal logistic performance, inventories should be minimized; in fact, the best scenario for a company is when they are not existent at all. In 10 (83.3%) firms, there is a precise knowledge of the location of all different stored materials; in 6 (50%) firms there exist procedures for stored items, that take into account the product characteristics: perishable goods, dangerous, and so on. 9 (75%) firms manage the location of materials by family and rotation; 6 (50%) firms have docks for loading / unloading and maneuvering areas: depending on the size and number and tonnage of vehicles that must be simultaneously received, 8 (66%) surveyed firms know exactly the cost of stored materials. In 11 (91.7) firms inventory levels of all products match their demand. 7 (58.3 %) firms have policies to reduce stocks in the entire chain. In 9 (75%) firms, obsolete materials are easily identified and monitored through a computer system. Even though that the results of this survey regarding logistic practices in inventory are encouraging, we cannot say they are the best that can be made, much remains to be done in this area. The final part of the survey is transport, and it is here where major problems are reported. Despite 10 (83.3%) firms know exactly the cost of transport, 8 (66.7) use the full vehicle capacity and design routes in order to reduce empty return trips. Only 5 (41.7%) firms reported a low rate of damage along transportation. These firms achieve this low rate due to proper product packaging and cargo transport standards. Moreover, only 7 (58.3%) firms reported high reliability in transport services as regards to time delivery with no major 9


failures. Strikingly enough for this leading production chain, 7 (58.3%) firms do not use data exchange systems with logistics service suppliers. Only 6 (50%) firms use route planning with communication systems such as Standard Short Messages (SMS) and Global System for Mobile Communications (GSM). Regarding the ability to distribute any lot size, 7 (58.3%) firms reported total flexibility.

The results from the original survey show the needed data to design a new supply chain strategy, such as the administration of a logistics network focused on the conditions of the Mexican infrastructure, regarding the number of locations, capacity and technology of manufacturing plants and warehouses, the selection of suppliers, assignment of product ranges for unity of space available in each transport vehicle, the selection of distribution channels and transportation modes, the flows of raw materials, semi-finished and finished products through the network. This paper finally introduces a novelty integrated scheme to manage the logistics network design problem in this country, according the restrictions for sustainable development.

10


References.

o Álvarez, M. (2002), “Cambios en la industria automotriz frente a la globalización: el sector de autopartes en México,” in Contaduría y Administración, Julio, Nr 206, Universidad Nacional Autónoma de México, México, pp. 29-49. o Bancomext

(2010),

Estados

e

indicadores

financieros,

Portal

Bancomext,

http://www.bancomext.com/Bancomext/portal/portal.jsp?parent=6&category=3598&d ocument=3445 march 2010. o Betanzo E. (1990), Plates-formes Logistiques: Analyse et Perspectives d'Implantation au Mexique, Thèse de doctorat en logistique, Université de Aix-Marseille II. o Caldeira, M., Nakano D. (2009), Knowledge and information flows in supply chain: A study on pharmaceutical companies, International Journal of Production Economics, Vol. 122, pp. 376 -384. o Hanjoul P., Thiesse F. (1983), Localisation de la firme sur un réseau, Localisation et Transports, Révue Economique, No.1, janvier, Paris. o Özbayrak M., Papadopoulou TC., Samaras E. (2006), A flexible and adaptable planning and control system for an MTO supply chain system, Robotics and Computer-Integrated Manufacturing, Vol. 22, pp. 557-565. o May A., Carter C. (2001), A case study of virtual team working in the European automotive industry, International Journal of Industrial Ergonomics, Vol. 27, PP. 171-186. o Sarimveis H., Patrinos P., Tarantilis C., Kiranoudis C. (2008), Dynamic modeling and control of supply chain systems: A review, Computers & Operations Research, Vol. 35, pp. 3530 – 3561. o Vaart T., Donk DP. (2008), A critical review of survey-based research in supply chain integration, International Journal of Production Economics, Vol. 111, pp. 42-55. o Vicencio, A. (2007), “La industria automotriz en México. Antecedentes, situación actual y perspectivas,” Contaduría y Administración, January, Nr. 221, Universidad Nacional Autónoma de México, Mexico, pp. 211-248. o Weber A. (1928), Theory of the location of industries, University of Chicago Press. o Wegner, S., Bode, C. (2006). An empirical investigation into supply chain vulnerability. Journal of purchasing and supply chain management. Vol 12, 312 pp.

11



Towards Sustainable ‘Take-back’. A Case Study of Retail Waste Collection Strategies. Maria Triantafyllou Transportation Research Group, University of Southampton, United Kingdom [email protected]

Dr Tom Cherrett Transportation Research Group, University of Southampton, United Kingdom [email protected]

In recent years, greater public awareness, stringent waste regulations and tight economic conditions have impacted on the development of traditional commercial waste service plans. Many businesses outsource their waste collection requirements to specialised third-parties whilst large retail chains often coordinate their waste management activities across their branch networks by centralising aspects of their collection system and placing additional national contracts with waste contractors. This paper uses a shopping centre as a case study example to identify the considerable logistics savings that could be gained through the establishment of centrally coordinated cross-supply chain waste service plans.

Key words: Waste Logistics, Reverse Logistics, Shopping Centre, Supply Chain Management, Retail Logistics

1


1 : INTRODUCTION Over the last decade there has been a substantial movement of retail systems from stand-alone units and clusters to more sophisticated integrated retail structures. The evolution of large shopping centres apart from offering more viable asset class for investors (in terms of real estate) and better accommodation for occupiers and consumers (in terms of more organised services provided) has also increased the opportunities for retailers, suppliers and shopping centre managers to create strategic alliances and obtain mutually beneficial goals in areas of common interest such as waste management operations. Potential rival businesses operating as part of an organised shopping centre have more opportunities to group together, share the burden and exploit any synergies that might arise from using an interlinked network for waste collections, rather than a series of disparate operations (Gardner, 2005; Elliot & Wright, 2004).

Retailers are now duty bound to improve their environmental performance through various legislation such as the ‘Duty of Care’, the ‘Landfill Tax’ and the ‘Packaging Waste Regulations’ (Pitt, 2005). In response, many businesses develop de-centralised reverse mechanisms to manage waste outputs in order to fulfil their legal obligations. Such practices often lead to little coordination in terms of treatment processes and the numbers of contractors used by businesses. Several studies (e.g. Fernie & Hart, 2001) have also revealed the difficulties smaller retailers experience trying to meet their responsibilities due to increased cost and workload. A more effective coordination of separate small-level waste flows could be attained through centrally coordinated waste management procedures and reverse logistics operations.

This paper aims to provide insights into the overlooked element of waste logistics and the different formations currently existing in the retail sector and more specifically in the context of a typical shopping centre in the UK. The main objectives of this paper are to: •

review the development of shopping centres and identify their considerable share in the total retail turnover.

•

identify the main legislative pressures that have brought waste logistics to the forefront of supply chain strategies. 2


•

identify the main waste collection systems currently existing in the retail sector and more specifically in the context of a shopping centre in the UK.

•

use a case study example to research the main characteristics of collective waste logistics strategies.

•

discuss the logistics implications of typical waste collection practices in the case study example.

•

generalise and compare the main findings to identify best practice examples and the opportunities of the retail industry especially in the context of a shopping centre to drive innovation into the integration and coordination of system’s wide physical flows.

2 : BACKGROUND 2.1: The Evolution of Shopping Centres in the Commercial World In recent years, a large portion of total retail space is taken by shopping centres (Ghisi et al, 2008). In the late 1990s, central Europe experienced a retail development boom driven by international hypermarket operators and speculative developers (Kok, 2007). Today, despite the current slowdown in shopping centre development due to the global downturn, many shopping centre projects across Europe receive the necessary permits and finance for construction, extension and renovation works. In the first half of 2009, 115 new shopping centres (3.1 million m2) opened across Europe, increasing the total shopping centre space to 123.7 million m2 (Marketbeat, 2009). In England and Wales, as of the first half of 2008, the total retail floor space accounted for 106.3 million m2 (ONS, 2008), with high street outlets and retail parks (1,340 parks) accounting for 55.3 million m2, shopping centres (819 establishments) for 12.3 million m2 and finally outlet stores, retail offices and other completing the existing retail landscape (BCSC, 2009).

2.2: Impacts of Legislation on Commercial Waste Management Operations Ongoing amendments to the regulatory framework exert significant pressures on retailers to follow changes and decide how to institute environmental decisions as part of their valueadding activities. In 1994, the ‘Packaging and Packaging Waste Directive [94/62/EC]’ was introduced with the aim to reduce packaging and institute more sustainable means to manage waste. A step change in the way commercial waste had to be treated and disposed of was the ‘Landfill Directive [99/31/EC]’ which set demanding targets to reduce the amount of waste 3


ending up in landfill. This has been followed by a set of legislative instruments such as the ‘Waste Electrical and Electronic Waste Directive (WEEE) [2002/96/EC]’ aiming to promote the reintegration of used products into productive processes and reduce their impact on the environment. Today, commercial premises have a ‘Duty of Care’ to make satisfactory arrangements for waste management and an obligation to set-up and manage material take-up schemes. 2.3: Commercial Waste Collection Systems Within the retail sector, two main mechanisms for waste management have been identified (Halldórsson & Skjøtt-Larsen, 2007). In a centralised collection model for waste and/or recyclables a single planner or organisation is acquainted with all system information including production levels, transportation capacities, associated sales prices, processing capabilities and disposal/treatment options for end-of-life products (Hong et al., 2007). Waste products derived from many different retailers are grouped in one place to enable efficient collection, inspection, disposition and redistribution. In contrast, a de-centralised collection system for commercial waste consists of multiple organizations involved in collection, sorting and distribution of returned items. It may be characterised by a vast amount of waste contractors, recyclers and other licensed logistics providers servicing individual businesses operating in an urban retail setting such as a commercial high street or a shopping centre.

2.4: Waste Management Practices in the UK Retail Sector Current market trends and existing social, legislative and economic conditions have direct impacts on the development of retail waste management practices. In the UK, commercial waste is traditionally collected individually by businesses through waste contractors. This method enables businesses to select a commercial service that best fits their operational, financial, legal and environmental requirements (Maynard & Cherrett, 2010). However in shopping centres one can identify a blend of different business types and sizes, different consumer flows, alternative disposal methods and existence or not of recycling plans that may impact on the level and diversity in waste production. (Pitt, 2005).

Today many shopping centre managers put in place centralised waste collection and disposal systems to facilitate and improve coordinated opportunities for the re-use and recycling of waste produced by tenants. In those cases the obligations and the commitments of landlords 4


and occupiers are regulated by specific green clauses introduced in standard lease contracts. Most of the shopping centres offer waste collection and recycling services for general waste, cardboard and plastic, while in many cases individual businesses recycle more waste streams. In most of the cases these businesses belong to large chains that have set out corporate protocols to increase control and coordinate waste management activities over their branch network. This can be achieved by centralising and formalising waste collection systems and placing national contracts with waste contractors and recyclers. Waste collections can be made directly from stores or through other consolidation centres using in-house or contracted fleet back-loading waste materials or servicing the retail establishment on a dedicated basis. Such strategies may co-exist within a shopping centre’s environment resulting in the development of hybrid waste collection systems. There is still no clear evidence about the logistics implications of such individual practices.

2.5: Need for Research Although several studies have largely researched the implications of various logistics formations in terms of products distribution (e.g. Dröge & Germain, 1989; Fernie, 1992; Eglese et al, 2005; Kohn & Brodin, 2007) or products collection for reuse and recovery (e.g. Fleischmann et al, 2000; Kusumastuti et al, 2004; Halldorsson & Skjott-Larsen, 2007; Mollenkopf et al, 2007) and while the literature presents a range of controversial arguments, little has been written about the physical design of different logistics networks for recycling or disposal. Most of the research has focused on the examination of single-product reverse supply chains with WEEE (e.g. mobile phones) attracting most of the interest (e.g. Lehtimen & Poikela, 2006; Chan & Chan, 2008; Ketal et al, 2008; Shih, 2001). Other less discussed topics refer to the implications of waste logistics practices for pharmaceutical waste (e.g. Ritchie et al, 2000), construction waste (e.g. Barros et al, 1998) and end-of-life vehicles (e.g. Cruz-Rivera & Ertel, 2008). Relatively little research has been done on the area of mixedwaste collection networks in a retail environment (e.g. Maynard & Cherrett, 2010) as most of the focus is still given to household waste collections (e.g. Jahre, 1995).

3 : CASE STUDY In order to give an in-depth insight into the physical designs of different logistics networks for recycling or disposal in a retail context the study uses a dedicated shopping centre in Hampshire, UK. The centre incorporates 2 department stores and 94 retail and catering units 5


selling a variety of goods including clothing, footwear, electrical products, jewellery, games, sports goods, books and optical goods. As part of the standard tenant contract, retailers are committed to maximise recycling and develop long-term waste minimisation strategies. The 2 department stores and one retail outlet have developed in-house waste collection mechanisms while 93 businesses have joined a centrally coordinated waste collection and disposal service for recyclable materials including cardboard, polythene, glass, paper, coat hangers, pallets, cages/scrap metal, fluorescent lighting tubes and wood. In addition a number of retailers produce hazardous wastes such as WEEE, batteries, clinical waste, fluorescent lighting tubes and used cooking oil, being either back-loaded to head offices, centralised distribution centres or other process/treatment plants. This is done using in-house collection systems, predominantly business specific, using in-house fleets or outsourced collections via thirdparty logistics providers or specialised waste contractors.

A review of the waste management procedures across the retailers operating in the shopping centre identified four main waste collection systems (Figure 1). Shopping Centre Waste Management Procedures

All waste products collected centrally by the shopping centre.

General waste collected centrally by the shopping centre and recyclate back-loaded via delivery vehicles.

General waste collected centrally by the shopping centre and recyclate collected by waste contractors.

All waste products back-loaded via delivery vehicles.

Figure 1:Waste collection systems identified in the case study shopping centre.

4 : DATA COLLECTION PLAN Waste operations were quantified as part of a wider data collection exercise researching the forward and reverse logistics activities used by businesses in the centre for core goods, product returns and waste/recyclate. A two-stage interview process was adopted including face-to-face contacts with the management team of the shopping centre and the managers of the retail outlets and catering units followed by phone interviews with key actors including third-party logistics providers, waste contractors, recyclers and head offices dealing with the 6


collection, transport, treatment, recovery or disposal of general waste, hazardous materials and recyclables. In addition, further information about the processes used by the retail companies, the logistics providers and the waste contractors was gathered from on-line sources.

The first survey stage employed a structured interview, incorporating qualitative and quantitative questions based on other recent urban freight survey methodologies (DfT, 2004; TRG, 2006) to research the treatment/handling practices developed on-site by individual businesses, the capacity of the systems in place (labour, equipment, storage space), the waste arisings, product testing ‘gate-keeper’ issues, collection contracts and origin-destination data for the associated outbound logistics activities. During the first survey stage, a 96% response rate was achieved eliciting detailed responses from 92 retail outlets in the shopping centre. In addition a number of selected case studies were further researched to obtain a full picture of the range of the formations for waste collections currently existing in a typical urban shopping centre in the UK.

5 : RESULTS 5.1: The Case of the Centralised Waste Collection System The study identified that the shopping centre’s management team has a lead role in working with tenants to incorporate principles of sustainable waste management. Through the provision of guides, programmes and training sessions, tenants are given the opportunity to reduce waste production and comply with environmental legislation. As part of the standard lease contract tenants are formally bound to increase environmental accountability into dayto-day operations and maximise recycling so that 80% of their waste stream is recovered. In addition they are expected to develop effective long-term waste minimisation strategies, reduce the use of single-trip packaging including carrier bags, and replace packaging materials with reusable, recyclable or compostable alternatives wherever possible.

The managers of the retail complex have developed an efficient in-house waste management system joined by 93 tenants. The shopping centre holds certificates for General Waste disposal and has the right to collect, treat and dispose of retailers waste under an agreed service charge. Waste collected is initially consolidated at 28 collection points at the back of the stores or in designated areas in the stock rooms in the case of the larger stores. The 7


shopping centre provides trolleys, pallets and cages to assist retailers with the waste collection. General waste is disposed of in 1100lt bins and sacks with recyclables such as polythene bags used for packaging and flattened cardboard boxes are disposed of in roll cages for recyclate located in the 28 designated collection points. The shopping centre owns two vehicles which empty the bins and cages twice a day and move waste into one of the three designated service yards (A, B and C). The use of the service yards for the waste collections is based upon the waste type, the equipment available and the easiness of access for waste contractors. General waste is consolidated in all service yards where static (A and C) and mobile (B) compactors are used to minimise waste volumes. Balers are located in service yard C and used to compress recyclables such as cardboard, paper and plastics into compact bales to enable handling, transport and storage. The majority of hazardous wastes generated on site (e.g. chemicals, paints and solvents) are managed by the cleaning contractor and are collected from service yard A which offers easier access to drivers. WEEE along with end-of-life fluorescent lighting tubes produced on site is collected from service yard C.

In 2007, the shopping centre recycled over 434 tonnes of waste, while the remaining waste was sent to a regional Energy Recovery Facility and used to produce electricity to power around 100 homes. General waste, mixed paper and hazardous waste produced on site are collected by ‘Waste Contractor A’. The collection frequency varies from weekly collections for general waste consolidated in service yards A and B to fortnightly collections for general waste consolidated in service yard C. Paper bales are collected on a monthly basis while hazardous waste on a weekly basis. Waste is shipped to the local incinerator (Marchwood, Southampton). In addition the centre holds a contract with ‘Waste Contractor B’ for the collection of cardboard sent to a recycling facility in Kent when 42 bales have been consolidated (approximately every month). Polythene and other plastic materials such as broken hangers produced by stores selling clothing are collected by ‘Waste Contractor C’ and shipped to a recycling facility in Derby every two months.

5.2: The Case of the Centralised Return Centre for Waste Business A belongs to a network of 375 stores in the UK trading a range of sports and leisure merchandise. The Group runs a fleet of 40 trailers and over 130 company cars travelling more than 8.7 million kilometres a year. The major source of waste production in the Group is packaging used to protect products in transit. In 2008 the amount of landfilled waste 8


amounted 1,400 tonnes. In attempting to reduce its carbon footprint, energy consumption and waste the Group has developed a centralised waste collection system in its branch network to recycle waste paper, cardboard, plastic and occasionally hazardous materials such as ink toners, redundant IT equipment and fluorescent lighting tubes. In 2008 the Group recycled 60 tonnes of waste paper, more than 5,500 tonnes of cardboard and 860 tonnes of baled plastic as a result of stores backfilling their waste to the National Distribution Centre located in Nottingham. Following the energy consumption at stores the fleet is the second largest producer of CO2 within the Group. Unnecessary mileage is avoided by not having fixed delivery/collection routes but instead optimising route planning to ensure that the full capacity of trailers is always used. Finally the Group provides energy training to all its employees as part of their employment induction and awards stores achieving specific recycling targets. Business A occupies one of the largest retail units in the shopping centre. The great variety of sports and leisure products offered to customers increases the need for storage and process capacity and has direct impacts on the volume and variety of waste/recyclates produced. Business A has joined the Group’s national waste collection system and therefore has separate equipment (e.g. roll cages) from the rest of the retail outlets in the shopping centre. General waste, mixed paper, cardboard, polythene used for packaging and plastic (e.g. broken hangers) are consolidated in a designated area in the stock room and left for collection at the back of the store. Waste and recyclables stored in plastic bags and loaded in roll cages are shipped back to the National Distribution Centre using the delivery fleet. Business A receives 5 deliveries a week during standard trade periods and 7 deliveries during busy trade periods (e.g. Christmas) as part of ‘milk-run’ delivery trips calling in another three regional retail outlets. The service order alters per day of the week to enable waste collections from the last stop. This technique enables the usage of the full load capacity for waste collections without mixing forward and reverse flows. Despite its benefits, the efficiency of the system is minimised due to the load constraints in place because of the need to back-load reusable packaging and other equipment (e.g. roll cages, pallets and plastic totes). This adds additional burden on Business A especially during busy trading periods when service efficiency and storage capacity are determinants for the business efficient operation.

5.3: The Case of De-Centralised Waste Collections Many retail outlets and catering units produce a range of waste products such as hazardous materials that require managed return and processing. Stringent waste legislation along with 9


other drives such as privacy issues needed to maintain businesses market position dictate specific handling, control and separation from the principal waste streams to ensure safe and efficient collection, transportation, treatment and disposal. In response many businesses have developed distinctly different handling protocols, treatment practices and disposal alternatives resulting in a multiple contract waste management regime with direct impacts on waste logistics operations (Triantafyllou & Cherrett, 2009a).

During the course of the data collection task it was identified that in total, 19 stores were producing hazardous waste as part of their commercial activities. All of them were obligated under several pieces of legislation to make special arrangements for the safe collection, storage and disposal of hazardous products. The surveys suggested that 3 restaurants were producing used cooking oil and another store selling optical equipment was producing clinical waste (e.g. needles, medicines, chemicals). In addition 11 businesses selling a variety of electronic products (e.g. computers, printers, television sets, game consoles and mobile phones) had in place individual collection arrangements. These businesses were obligated under the WEEE Directive to develop customer take-back schemes at stores. Six stores producing WEEE were also producing batteries and used the same mechanisms for their collection. Finally another 4 stores were producing batteries collected either by the shopping centre or back-loaded/posted to head offices.

It was estimated that during a typical (not busy) week, the retail outlets and catering units in the shopping centre produced in total approximately 11 totes filled with WEEE and batteries, 1 sharp box and 1 bag filled with clinical waste and 1 drum used to store used cooking oil. Although hazardous waste was found to be only a very small proportion (0.01%) of the total waste produced (approximately 3,900 items of any size) the demand for transport was quite significant with 15 dedicated and 6 collection trips being made in the shopping centre during a typical week. These service levels appeared to be significantly high when compared to the ones of the centralised waste collection system.

5.3.1 : The Case of De-Centralised Waste Collections for Mobile Phones The study identified that 4 out of the 5 stores selling mobile phones had employed the same waste contractor to collect end-of-life devices, while the fifth business had developed inhouse mechanisms (fleet, treatment/process activities) to manage the return flow of mobile 10


phones. Although the 4 businesses outsourced collection, treatment and disposal services to the same third party contractor (3PL) the collection services varied significantly. In more detail 2 businesses provided their customers with pre-paid envelopes to ship mobiles via the Postal network, another received the services of a 3PL whenever requested, while the latter used a 3PL’s delivery vehicles to ship handsets back to the waste contractor’s processing hub in Essex every week.

A further examination of the individual waste collection practices in the above example brought out several operational complexities impacting on the efficiency and effectiveness of each system. In the first case, the use of the existing postal network enabled a quick customer response without placing any burden on the existing logistics network. However, in a comparison of drop-off and mail-in collection methods, Hanafi et al. (2008) concluded that due to the small size of mobile phones, the cost of mailing each product to the transfer/consolidation station can be quite high in contrast with the cost of the drop-off method. In the second case, the system’s ability to react quickly to emergency collections was counteracted by the lack of integration of individual logistics practices into a ‘network system’ in terms of using the load capacity in the inbound leg of the collection trip. Finally, in the third case, the utilisation of delivery vehicle spare capacity to carry loads on the return journey was minimised by load restrictions set by the 3PL, inhibiting the collection of end-oflife products over a set volume level. This has meant that the business should either arrange to receive a rather costly emergency collection or keep and store end-of-life goods in its valuable retail space. This could potentially create negative safety and branding issues especially in the case of mixing in-bound and out-bound flows of goods.

6 : DISCUSSION The examination of the principal waste collection systems currently existing within the case study shopping centre can provide enough evidence for the identification of the main characteristics of typical waste service plans in the retail sector.

6.1: Centralised Waste Collection Systems In an integrated retail structure there are increased opportunities to coordinate centrally activities related to the collection of products discarded by a group of retail outlets (Figure 2). Its application enables high consolidation of waste at source and therefore results in reduced 11


service visits and logistics footprint. Its success depends on the participation and synchronisation of retailers and requires clear definition of the rights and obligations of the landlord and the tenants through a consistent set of rules, regulations and contracts. The role of a ‘logistics controller’ is also very important as this person is responsible for the coordination of products and waste take-back using the delivery fleet or 3PLs in such a way to minimise the number of overall freight vehicle trips, the distance travelled and the number of less-than-vehicle loads generated (Triantafyllou & Cherrett, 2009b). This central figure is also responsible for the integration of corporate strategic planning into day-to-day processes by looking beyond the organisation’s processes and find and eliminate sources of waste to optimise collections (Melnyk et al, 1998). This person must identify the need of resources in terms of capital structure and human information, develop the technological and structural capabilities, modernise processes and actions and evaluate performance to identify the required adjustments (Ghisi et al, 2008).

Integrated Retail Structure Central Planner

Retail Business A

Retail Business B

Retail Business C

Retail Business D

Waste Stream A

Waste Stream A

Waste Stream A

Waste Stream A

Central Collection Waste Stream A

Waste Stream B

Waste Stream B

Waste Stream B

Waste Stream B

Central Collection Waste Stream B

Waste Stream C

Waste Stream C

Waste Stream C

Waste Stream C

Central Collection Waste Stream C

Waste Stream D

Waste Stream D

Waste Stream D

Waste Stream D

Central Collection Waste Stream D

Figure 2: Centralised waste collections within an integrated retail structure.

6.2 Centralised Return Centres for Waste The use of centralised return centres (CRCs) to process waste materials and recyclables collected through a business branch network (Figure 3) is usually adopted by large retail chains having the capital to make investments into processes that integrate environmental management into corporate strategic planning and resource productivity frameworks. This helps businesses increase their competitive advantage in a socially responsible manner. In 12


logistical terms, this system utilises the backload capacity of delivery fleet and therefore minimises the transport footprint. In addition the savings from having fewer premises usually outweigh the additional costs of transporting goods longer distances (McKinnon et al, 2002). However CRCs operation has many implications in terms of time and workload management both for staff involved with the management of waste on-site and store managers being responsible for the organisation of waste take-back using in-house/contracted delivery or 3PL fleet. The organisation of collection activities must be made in such a way to prevent waste from occupying valuable retail space and without impeding the capacity of the system to deliver products in follow-up destinations or interfering with the capacity of the system to process product returns. Coupled to these issues, the lack of equipment at stores used to reduce waste volume prior collection, the need to transport re-usable or one-off packaging to ship waste, as well as the inability of the system to manage emergency collections are some of the issues a business should consider before embracing this type of waste collection technique.

Integrated Retail Structure Business B Branch A

Retail Business A

Retail Business B

Waste Stream A

Waste Stream A

Business B Branch B

Centralised Return Centre for specific Waste Streams

Business A Branch A

Waste Stream B

Waste Stream B

Waste Stream C

Waste Stream C

Dedicated Waste Collection

Waste Stream D

Waste Stream D


Business A Branch B Business A Branch C

Centralised Return Centre for all Waste Streams

Figure 3: Centralised return centres for a) all waste streams (Business A) and b) specific waste streams (Business B).

13


In such a system one can distinguish between different coordination arrangements. For example the CRCs for waste collections may differ from the CRCs for products returns where employees assess the condition of each incoming item, and determine the best place to dispose of the item (gate-keeping operations). In addition forward and reverse distribution centres may differ because reverse CRCs are often more profitable for companies when outsourced (Beltran, 2002). On the other hand, a number of firms prefer the use of local hubs that allow local producers to pool their resources to reduce costs, cut carbon emissions and lower the overall environmental impact of product distribution and waste/returns collections. When products and waste are managed in the same centre then significant problems may arise in association with the mixing of the two flows. In those cases operations related with the management of products to be delivered must precede the operations related with the management of waste. Finally locating a CRC for distribution or product returns has been object of extended research with much of the focus being given on the optimisation of the location of hubs. Often researchers employ fuzzy goal programming in their models to examine issues such as capacity (Nie et al, 2009) and load allocation (e.g. Mujumdar & Vemula, 2004). In determining the siting of a CRC for waste and recyclables one should consider the local/regional recycling/disposal network along with the business branch network in order to eliminate the overall distances travelled. The optimal solution can be attained by case-by-case examination of the problem.

6.3: De-Centralised Waste Collections The development of the existing multiple contract waste management regime may result from absence of centralised waste collection arrangements on site or over the branch network. In addition the great variety in material characteristics and properties of waste/recyclates often dictate specific handling and treatment measures (control and separation) from the principal reverse waste flow (e.g. hazardous waste materials). Conflicting priorities, individual requirements including different collection time windows, incompatible waste types, different handling requirements, branding and privacy issues impeding retailers from exchanging information required to support integrated decision making, are some of the reasons why the retail industry has failed to drive innovation into the integration and coordination of system’s–wide physical flows. On the other hand, McLeod et al, (2007) concluded that when de-centralised returns networks take place, the existence of competent inspection, evaluation

14


and disposition mechanisms for returns are essential in order for costs and vehicles kilometres to be minimised (Figure 4). Integrated Retail Structure

Retail Business A

Retail Business B


Waste Stream A

Waste Stream A



Waste Stream B

Waste Stream B



Waste Stream C

Waste Stream C



Waste Stream D

Waste Stream D


Figure 4: De-centralised waste collections within an integrated retail structure.

7 : CONCLUSIONS The study identified that the considerable share of shopping centres in the total retail turnover and the strong presence of large retail chains have impacted on the formation of traditional centralised and de-centralised waste management practices with businesses striving to meet their waste management obligations while keeping their competitive advantage. Through a case study example, three main waste collection models were identified. The study concluded that within an integrated retail structure there are increased opportunities to develop centralised waste collection mechanisms. The efficiency and effectiveness of this system requires central coordination from an individual/body that is responsible for integrating corporate strategic planning into day-to-day waste management processes, finding and eliminating sources of waste to optimise collections and identifying the need for resources in terms of capital structure and human resources. Such a central co-ordinator would also have to encourage tenants to buy into the scheme, monitor their performance, coordinate the takeback of waste using in-house or 3PL fleets and research the local/regional opportunities to treat and dispose of waste/recyclables to minimise the overall logistics footprint of the centre. The comparison of centralised waste collection with CRCs for waste provides strong evidence for the ability of the former to not only bring together a high number of low volume flows and 15


enable consolidation, but to also minimise the uncertainty of the system and therefore the number of inefficient emergency collections. In addition, considerable savings in terms of time, labour effort and retail space could be achieved through the introduction of formal agreements between shopping centre landlords and tenants and the transfer of tenant responsibilities to shopping centre managers. Finally, the apparent failure of the decentralised waste collection systems to match low waste production with the number of contracts and the associated logistics activity suggests that there could be considerable potential for a centralised system to integrate and effectively coordinate separate small-level waste flows.

8 : REFERENCES Barros A.I., Dekker R., Scholten, V. (1998), A Two-Level Network for Recycling Sand: A Case Study, European Journal of Operational Research, Vol. 110, No2, pp. 199-215.

BCSC (2009) Retail Property Statistics. Summary, BSCS, October [online] Available from: www.bcsc.org.uk [Accessed: 13 February 2010]

Beltran L.S. (2002), Reverse Logistics: Current Trends and Practices in the Commercial World, Logistics Spectrum, Vol. 36, No3, pp. 4-8.

Chan T.S.F., Chan K.H. (2008), A Survey on Reverse Logistics System of Mobile Phone Industry in Hong Kong, Management Decision, Vol. 46, No5, pp. 702-708.

Cruz-Rivera R., Ertel J. (2008), Reverse Logistics Network Design for the Collection of Endof-Life Vehicles in Mexico, European Journal of Operational Research, Vol. 196, No3, pp. 930-939.

DfT (2004), The Efficiency of Reverse Logistics, Cranfield School of Management, Sheffield Hallam University and the Chartered Institute of Logistics and Transport.

Dröge C., Germain R. (1989), The Impact of the Centralized Structuring of Logistics Activities on Span of Control, Formalization and Performance, Journal of the Academy of Marketing Science, Vol. 17, No1, pp. 81-89. 16


Eglese R.W., Merger A., Sohrabi B. (2005), The Grocery Superstore Vehicle Scheduling Problem, Journal of the Operational Research Society, Vol. 56, No2005, pp. 902-911.

Elliot M., Wright J. (2004), The Future Direction of Supply Chain: Mastering Reverse Logistics, Whitepaper, Accenture, June.

Fernie J. (1992), Distribution Strategies of European Retailers, European Journal of Marketing, Vol. 26, No8/9, pp. 35-47.

Fernie J., Hart C. (2001), UK Packaging Waste Legislation - Implications for Retailers, British Food Journal, Vol. 103, No3, pp. 187-197.

Fleischmann M., Krikke R.H., Dekker R., Flapper P.D.S. (2000), A Characterisation of Logistics Networks for Product Recovery, Omega, Vol. 28, No6, pp. 653-666.

Gardner S. (2005), Reverse logistics–A forgotten necessity? Manufacturing & Logistics IT [online].

Available

from:

http://www.logisticsit.com/absolutenm/templates/article-

critical.aspx ?articleid=744&zoneid=31 [Accessed: 19 November 2008]

Ghisi A.F., Da Solveira A.G.J., Kristensen T., Martin H., Hingley A., Lindgreen A. (2008), Horizontal Alliances amongst Small Retailers in Brazil, British Food Journal, Vol. 110, No4/5, pp. 514-538.

Halldorsson A., Skjoett-Larsen T. (2007), Design of Reverse Supply Chains: Centralized or Decentralized Structure, in Delfmann W., de Koster R. (Eds.), Managing Supply Chains Challenges and Opportunities. Copenhagen Business School Press. pp. 1-26.

Hanafi J., Kara S. & Kaebernick H. (2008), Reverse Logistics Strategies for End-of-Life Products, The International Journal of Logistics Management, Vol. 19, No3, pp.367-388.

Hong I-H., Ammons C.J., Realff J.M. (2008), Centralised versus De-Centralised Decision Making for Recycled Materials Flows, Envir. Sci. Technol., Vol. 42, No4, pp. 1172-1177. 17


Jahre M. (1998), Household Waste Collection as a Reverse Channel. A Theoretical Perspective, International Journal of Physical Distribution & Logistics, Vol. 25, No2, pp. 3955.

Ketal H., Li L., Wenying D. (2008), Research on Recovery Logistics Network of Waste Electronic and Electrical Equipment in China. Proceeding of the 3rd IEEE Conference on Industrial Electronics and Applications, Singapore 3, 4 and 5 June.

Kohn C., Brodin M.H. (2007) Centralised Distribution Systems and the Environment: How Increased Transport Work can Decrease the Environmental Impact of Logistics, International Journal of Logistics Research and Applications, Vol. 1, No1, pp. 1-17.

Kok H. (2007), Restructuring Retail Property Markets in Central Europe: Impacts on Urban Space, Journal of Housing and the Built Environment, Vol. 22, No1, pp. 107-126.

Kusumastuti R.D., Piplani R., Lim G.H. (2004), An Approach to Design Reverse Logistics Networks for Product Recovery, in: Xie M., Durrani T.S., Tang H.K. (Eds.), Innovation and Entrepreneurship for Sustainable Development, Engineering Management Conference. IEEE, Singapore 18, 19, 20 and 21 October.

Lehtimen U., Poikela K. (2006), Challenges of WEEE on Reverse Logistics: A Case Study on a Collection Network in Finland, Proceedings of the Logistics Research Network (LRN 2006), Newcastle Upon Tyne, UK, 6, 7 and 8 September.

Marketbeat (2009), Shopping Centre Development Report Europe 2009, London, Cushman & Wakefield, September.

Maynard S., Cherrett T. (2010), A Study of Waste Logistics in the Retail Sector – Opportunities for ‘Greening’ the Take-Back Process, Proceedings of the 89th Annual Meeting of the Transportation Research Board (TRB 2010), Washington, D.C., USA., 10, 11, 12, 13 and 14 January.

18


McKinnon A., Marchant C., Somerville A. (2002), Freight transport and logistics in west central Scotland, Report to West of Scotland Transport Partnership, December.

McLeod F., Hickford A., Maynard S., Cherrett T.J., Allen J. (2007), Developing Innovative and More Sustainable Approaches to Reverse Logistics for the Collection, Recycling and Disposal of Waste Products from Urban Centres. Literature Review and Identification of Opportunities. Green Logistics Project, Transportation Research Group, University of Southampton, January. Available from: http://greenlogistics.org/SiteResources/9d09bda8-a985-4241-b4ce-bc6ddaf7 ea31_Reverse%20logistics%20report%20WM10.pdf [Accessed: 10 January 2008]

Melnyk S.A., Walton S.V., Handfield R.B. (1998), The Green Supply Chain: Integrating Suppliers into Environmental Management Process, International Journal of Purchasing and Materials Management, Vol. 34, No2, pp. 2-11.

Mollenkopg D., Russo I., Frankel R. (2007), The Returns Management Process in Supply Chain Strategy, International Journal of Physical Distribution & Logistics Management, Vol. 37, No7, pp. 568-592.

Mujumbar P.P., Vemula S.V.R. (2004), Fuzzy Waste Load Allocation Model: SimulationOptimization Approach, Journal of Computing in Civil Engineering, Vol. 18, No2, pp. 120131.

Nie X., Huang G., Li Y. (2009), Capacity Planning for Waste Management Systems: An Interval Fuzzy Robust Dynamic Programming Approach, Journal of the Air & Waste Management Association, Vol. 59, No11, pp. 1317-1330.

ONS (2008), Commercial and Industrial Floorspace and Rateable Value Statistics (2005 Revaluation) 2008, Office for National Statistics, Newport, UK [online]. Available from: http://www.neighbourhood.statistics.gov.uk [Accessed: 10 February 2010] Pitt M. (2005), Trends in Shopping Centre Waste Management, Facilities, Vol. 23, No11-12, pp. 522-533. 19


Ritchie L., Burnes B., Whittle P., Hey R. (2000), The Benefits of Reverse Logistics: The Case of the Manchester Royal Infirmary Pharmacy, Supply Chain Management: An International Journal, Vol. 5, No5, pp.226-234.

Shih L-H. (2001), Reverse Logistics System Planning for Recycling Electrical Appliances and Computers in Taiwan, Resources, Conservation and Recycling, Vol. 32, No1, pp. 55-72.

TRG (2006) MIRACLES 2006 Final Deliverable D4.2: Report on evaluation Results, Transportation Research Group, University of Southampton, U.K., August.

Triantafyllou M., Cherrett T.J. (2009A), The Logistics of Managing Hazardous Waste - A Case Study Analysis in the Retail Sector, Proceedings of the Logistics Research Network Annual Conference (LRN 2009), Cardiff, UK, 9, 10 and 11 September.

Triantafyllou M., Cherrett T.J. (2009b), Developing Business Establishment Surveys to Understand Reverse Logistics Processes within a Multi-Retailer Shopping Environment, Proceedings of the 88th Annual Meeting of the Transportation Research Board (TRB 2009), Washington D.C., USA., 11, 12, 13, 14 and 15 January.

20