ronald d. fitzmartin

Published in support of

and

|

Q1 2008

|

www.Futurepharmaus.com

Driving the industry forward

Tan Miller

Details, Details, Details

Shaping Success

Adaptive Clinical Trials Panel Moderated by Paul Gallo

PsyCROlogy of Patient Recruitment Panel Moderated by Peter DiBiaso

Raising the Bar

Supply Chain Panel Moderated by Mike Celentano

James C. Greenwood Fighting the Disease, Not the Cure

Eric Lightcap

RNAi Panel: Breaking New Ground

Novartis’ Alice Woychik Meets the Mark

Jim Phelan & Drew Phelan Like Father, Unlike Son

Cerner & Pfizer

Look at the Future of Drug Safety

On the Fast Track

Carestream Molecular Imaging

Taking a CloserLook at E-Detailing With closerlook’s CEO

Ronald D. Fitzmartin Strong and Growing DIA

[

]

Decision Support Systems:

A Detailed Approach Tan Miller, Regional Director of Distribution, Johnson & Johnson talks about why the details matter.

A

key determinant of whether the Decision Support Systems (DSS) for your firm’s supply chain provide the necessary insights to facilitate effective decision-making is whether your DSS evaluate decisions with the appropriate level of detail. Too much or too little detail can equally cloud a problem and lead to well-intentioned, but sub-optimal decisions. Further, the apppropriate detail required varies widely by the type of decision and the level in a planning framework of a particular issue. In a previous article in Future Pharmaceuticals (November 2007) we reviewed “general” planning and scheduling frameworks for distribution and supply chain management. In this article we focus on an important component of these frameworks; namely, assuring that the multiple DSS that support a supply chain offer the correct granularity of detail at each planning level.

Background

A complete framework for supply chain planning must address all planning and scheduling time horizons from the strategic two or more years to the tactical 12 to 24 months to the operational one to 18 months. See Miller (2007 and 2002) for a review of these frameworks. Each of these planning levels requires DSS support and the reader interested in DSS applications is referred to Liberatore and Nydick (1998) for a general review. To

illustrate the importance of identifying the appropriate level of detail, we will first consider a specific example and then draw inferences from this. Example: The Potential Use For Item Level Detail In Tactical Production Planning Applications

Let us consider a manufacturing firm that produces and sells several thousand unique items per year. Further, assume this firm is developing a production plan for its multi-plant manufacturing network for the next 12 to 18 months. The firm manufactures products that have the heirarchy displayed in Figure 1. Note in this figure that the firm’s overall set of products (level 1) disaggregates first into product lines (level 2), then into product families (level 3) and finally into unique end items (level 4). Product families represent aggregations of end items that have similar characteristics. For example, a group of end items that can be produced on the same production lines, at the same plants and at about the same cost and rate per unit represent a typical product family. Planning network-wide production levels by plant for a 12 to 18 month horizon represents an activity typically performed at a level of aggregation greater than the end item. It is generally not feasible or efficient to plan individually for several thousand

figure 1 illlustrative product aggregation hierarchy

90

Firm’s Total Product Line

Level 1

Major Product Lines

Level 2

Product Families of (Product Lines)

Level 3

End Items

Level 4

FUTURE PHARMACEUTICALS

DRIVING THE INDUSTRY FORWARD

supply chain & logistics

TABLE 1 product family level evaluation TOTAL PRODUCTION REQUIREMENTS OVER PLANNING HORIZON (When Evaluated at Product Family Level) (1)

(2)

(3)

(4)

Beginning Inventory Position

Annual Selling Rate

End of Planning Horizon Inventory Target

Total Production Required Over Next 12 Monthsa

9,200

16,000

3,200

10,000

a) calculation for column (4) = maximum of: [column (2) + column (3) - column (1), or 0]. *The Planning horizon in this example is the next twelve months. items in a manufacturer’s network for a 12 to 18 month planning horizon. Thus, product families or in some cases, higher levels of aggregation, represent the appropriate planing level for tactical capacity and production planning. When planning production capacity at product family or higher aggregate levels, however, it is important to understand if there are end item factors that must be accounted for in the initial planning process. The following tables provide a numerical illustration of a situation where initial analyses at the end item level must be completed prior to performing annual product family-based production planning. The firm determines its production requirements as: ■ Forecast Sales + Ending Inventory Target – Beginning Inventory For illustrative purposes, we assume that the firm’s product line of several thousand items aggregates into 75 product families. To plan production for these 75 product families, the firm

must develop beginning inventory numbers for each product family, as well as forecasts and ending inventory targets. Table 1 displays an analysis of the production requirements for one product family calculated at the product family level. Similar calculations would be performed for all 75 families. Note that this product family has 9,200 units in inventory at the beginning of the planning horizon, it has forecast sales of 16,000 units and an end of horizon inventory target of 3,200 units. Table 1 shows that with some rudimentary calculations we can determine that this product family requires 10,000 units of production over the planning horizon. Table 2 offers an analysis of the production requirements for this same product family calculated with one important difference. This figure displays the calculations that result when the beginning inventory of the product family is evaluated at the end item level, and then is aggregated to generate a beginning of period product family inventory number. Note that in this example the product family contains five end items. The

TABLE 2 end-item level evaluation TOTAL PRODUCTION REQUIREMENTS OVER PLANNING HORIZON (When Evaluated at end-item Level) (1)

(2)

(3)

(4)

(5)

(6)

(7)

End Item

Beginning Inventory Position

Annual Selling Rate

1

400

2,000

400

2,000

400

0

2

200

3,000

600

3,400

200

0

3

100

1,000

200

1,100

100

0

4

300

4,000

800

4,500

300

0

5

8,200

6,000

1,200

0

7,200

1,000

Total

9,200

16,000

3,200

11,000

8,200

1,000

End of Total Production Useable Excess Planning Horizon Required Over the Inventory Over Inventory Over Inventory Target Next 12 Monthsa Planning Horizonb Planning Horizonc

a) calculation for column (5) = maximum of: [column (3) + column (4) - column (2), or 0]. b) calculation for column (6) = maximum of: [column (2) + column (3) - column (4), or 0]. c) calculation for column (7) = maximum of: [column (2) - column (6), or 0]. *The Planning horizon in this example is the next twelve months.

www.futurepharmaus.com


91

decision support systems

figure 2 product family vs. end-item level evaluation INVENTORY EVALUATED AT:

PRODUCTION REQUIREMENT

USEABLE INVENTORY

EXCESS INVENTORY

Family Level

10,000

9,200

0

Item Level

11,000

8,200 •

1,000

Production requirements underestimate resulting if inventory is not evaluated at end-item level 10,000 8,000

8,200

6,000 4,000 2,000 0

1,000 7,200

Beginning Inventory Level

Inventory Target

1,200

Annual Sales

6,000

Total Inventory Needed Over Planning Horizon (i.e., Useable)

respective total beginning inventory, forecast annual sales and end of planning horizon inventory target of these five items sum to the identical numbers shown for this product family in Table 1. A quick review of the total row for columns (5) through (7) in Table 2 reveals though that a different perspective of the inventory position on this product family emerges at the end item level. Specifically, analyzing the beginning inventory at this level shows that the vast majority of the total current inventory of this product family resides in end item 5. Further, calculations show

“Planning network-wide production levels by plant for a 12 to 18 month horizon represents an activity typically performed at a level of aggregation greater than the end item.” that 1,000 units of this end item’s inventory are “excess over the planning horizon”; and therefore, only 8,200 units (column 6) of the inventory of the product family are “useable over the planning horizon.” For this illustration, we define the maximum “useable” inventory during a planning horizon as the forecast sales in units plus the ending inventory target. Additional inventory above this level is not required during the planning horizon and is defined as “excess” for this period. Assuming that there is not substitutability among end items within the product family, a firm in this example must actually produce 11,000 units of

92


Production Requirements Underestimate, If inventory is Only Evaluated at Product Family Level

Excess Inventory Over Planning Horizon

the product family during the planning horizon, rather than the 10,000 units calculated by the analysis conducted strictly at the product family level. Figure 2 summarizes the difference that will result in the example from a production requirements forecast based upon a beginning inventory number generated at the product family level and one generated by first evaluating inventory at the end item level. To plan production requirements accurately, the firm must recognize that it has 1,000 units of inventory in item 5, and therefore in item 5’s product family, which it cannot utilize over the planning horizon. For production planning purposes, including these 1,000 units in the product family’s beginning of period inventory figure will overstate the true level of inventory “useable over the planning horizon” for this product family. Thus, the quantity displayed in column (6) of Table 2 (8,200), rather than the quantity shown in column (5) of Table 1 (9,200), represents the proper amount to use as this family’s “beginning of period inventory” in an annual production/distribution planning model. For other reasons, (e.g., financial, special marketing programs, etc.), one must account for this product family’s excess inventory over the planning horizon in other planning areas. This example illustrates that evaluating product family inventory positions and production requirements at the product family level makes it impossible to recognize if there is any excess or unuseable inventory over the planning horizon in any of the end items used within the product family. By evaluating product family inventories at the more disaggregated item level and then summing end item results to obtain the beginning of period product family inventory data, one can accurately evaluate a product family’s current level of “useable inventory over the planning horizon.” The use of


supply chain & logistics

“By evaluating product family inventories at the more disaggregated item level and then summing end item results to obtain the beginning of period product family inventory data, one can accurately evaluate a product family’s current level of useable inventory over the planning horizon.” this hierachical approach assures that a firm will not, as a result of a mis-evaluation of the utility of its existing inventory, underestimate the true level of production capacity that it will require over a planning horizon. Implications For Decision Support Systems

The preceding example offers a quick illustration of how a firm must evaluate its DSS requirements. At the strategic, tactical and operational levels of planning, analyses and outputs are evaluated and decisions made based on different levels of detail. Managers should review their DSS to determine whether each application generates planning outputs based on the correct level of detail. Futher, it is not sufficient to simply review the output of a DSS to determine if it is appropriate. One must consider if additional detail is required in the intermediate analytic process to yield outputs that are ultimately reviewed at a higher level of aggregation. In the preceding production planning example, product families represented the correct level to formulate network-wide annual plant production capacity assignments. However, to obtain accurate product family production plans, we had to evaluate inventories first at the end item level. This example is based on the production planning process of an actual firm (see Miller, 2002). Until the firm expanded its annual production planning process to include an initial evaluation of inventory at the end item level, the firm consistently encountered customer service fill rate service issues. The IT technologies commonly in place in the industry today facilitate broad useage of these more advanced DSS analyses. Fifteen years ago, the capability to generate the detailed underlying analysis of the preceding production planning example might

have only resided in a firm’s IT department. Today, utilizing data warehouses and tools such as Access database querying etc, these analytic capabilities often reside in the non-IT user community of a firm as well. This expands the opportunity for managers to assure that their DSS outputs are based on appropriate detail. Further, the more traditional option to utilize the IT community of a firm always represents another option. The objective of this article is to encourage managers to evaluate their current DSS systems. We reviewed just one simple example of the “right” and “wrong” levels of detail for a production planning activity. Decision Support Systems for supply chains abound with similar examples. The reader is encouraged to reevaluate the DSS at their respective organizations. Well defined and correctly detailed DSS can provide significant competitive advantage. In closing, questions to consider include: ■ Are the appropriate DSS in place to support each level of your planning hierarchy? ■ Is the detail of analysis correct at each planning level and for each application? ■ Is there a need for a more hierarchical approach to some applications such as the production planning application we reviewed. For example, do outputs at one level require preceding analyses at more detailed levels in order to facilitate accurate “aggregated”outputs? FP References

Miller, T. C. (November, 2007), “Frameworks for Continuous Distribution and Supply Chain Performance Improvement,” Future Pharmaceuticals. Miller, T. C. (2002), Hierarchical Operations And Supply Chain Planning, Springer-Verlag, Heidelberg, Germany. Liberatore, M. L. and Nydick, R. (1998), Decision Technology For Business Applications, McGraw-Hill Companies, Primus Custom Publishing, Villanova, PA.

Tan Miller, Regional Director of Distribution at Johnson and Johnson Group of Consumer Companies, is currently responsible for the operations of the J&J’s U.S. Consumer Distribution Network. Previously he has held similar responsibilities for Pfizer, and he has also held production and distribution positions with Mercer Management Consulting, Unisys and American Olean Tile Company. Tan has published four books and over 50 articles on supply chain and logistics operations and planning.



93

Published in support of

and

FUTURE Driving the industry forward

pharmaceuticals

|

Q4 2007

|

www.Futurepharmaus.com

Robert abraham

vice president of oncology at wyeth research, see page 12

Katie Bronander

Planning a Green Meeting

Cornerstone of Success

Training & Learning Panel Moderated by Mike Capaldi

Plugging E-Pedigree

Supply Chain Panel Moderated by Doug McLeester

Sourav Kundu Integration of Single-Use Technologies

SGI

Scalable Energy

Biomarkers Mark the Way of the Future Moderated by Richard Frank

Kenneth Getz of CISCRP & Kathleen Drennan of Iris Clinical Trials and Tribulations

on the horizon a shift in the

pre-clinical paradigm

Frameworks for

Continuous Distribution and Supply Chain Performance Improvement By Tan Miller, Regional Director Distribution, Johnson & Johnson Group of Consumer Companies

In this article, we describe how a firm and its supply chain professionals can enhance their organization by maintaining robust frameworks of their current operations and planning processes. Frameworks provide a perspective for an organization’s managers to better understand their functions and processes. Further, with every passing year, new tools, techniques and methodologies become available for practitioners to utilize in both their short- term operations and their long-term planning activities. Well-maintained frameworks provide a vehicle by which to evaluate if and when new methodologies and decision support tools can improve the efficiency and effectiveness of a supply chain. We begin by reviewing a generic framework for supply chain planning and operations. Next, to illustrate that this general framework applies to many functions, we review the application of this generic framework to warehouse planning operations. Finally, we conclude with a discussion of the benefits of a framework approach. A Supply Chain Planning Framework

Figure 1 offers a general framework for supply chain planning that spans the strategic, tactical, and operational planning and scheduling levels. Importantly, we denote Figure 1 as a hierarchical supply chain planning framework (HSCP) because it includes bidirectional linkages between higher and lower level planning and scheduling activities. Beginning at the strategic level, a company must address such key issues as overall corporate objectives, market share and profitability goals, business and product mix, etc. Planning decisions on overall corporate objectives drive strategic supply chain decisions. For example, market share and business or product mix objectives will strongly influence manufacturing capacity strategies.

76


At the strategic manufacturing planning level, the company must address such issues as planned production capacity levels for the next three years and beyond, the number of facilities it plans to operate, their locations, the resources the company will assign to its manufacturing operations and numerous other important longterm decisions. Decisions made at the strategic level place constraints on the tactical planning level. Typical planning activities at the tactical level include the allocation of capacity and resources to product lines for the next 12 to 18 months, aggregate planning of workforce levels, the development or fine-tuning of distribution plans and numerous other activities. Within the constraints of the company’s manufacturing and distribution infrastructure (an infrastructure determined by previous strategic decisions), managers make tactical planning decisions designed to optimize the use of the existing infrastructure. Planning decisions carried out at the tactical level impose constraints on operational planning and scheduling decisions. At this level, activities such as distribution resouce planning, rough cut capacity planning, master production scheduling, shop floor control scheduling and many other decisions occur. The feedback loops from the operational level to the tactical level and from the tactical level to the strategic level represent one of the most important characteristics of the HSCP system illustrated in Figure 1. A true HSCP system is a closed-loop system that employs a “top-down” planning approach complemented by “bottom-up” feedback loops. Given the emphasis of HSCP systems on evaluating capacity levels and imposing and/or communicating capacity constraints from higher levels down to lower levels, it is imperative that strong feedback loops exist. As is well known, production and


Supply Chain & Logistics

distribution plans that appear feasible at an aggregate level can often contain hidden infeasibilities that only manifest themselves at lower, more disaggregated levels. Without the proper feedback loops embedded into a hierarchical planning system, the danger that a company will attempt to move forward with infeasible plans always exists. These infeasibilities often do not surface until a company is in the midst of executing its operational plans and schedules. The hierarchical framework presented in Figure 1 is generic in that, although individual HSCP systems may differ dramatically, most systems are designed within this or a similar general framework. Frameworks Have Broad Applications

One can utilize a hierarchical framework approach for virtually any major functional area within a supply chain. Transportation, inventory, demand management and warehouse operations represent just a few examples of major functional areas that can benefit from a hierarchical framework perspective.1 Furthermore, frameworks can also be applied to more general decision support activities such as performance measurement.2 To illustrate the broad applicability of frameworks, we now consider Figures 2 and 3, which offer brief illustrations of a framework for warehouse operations. Warehouse Operations

The warehouse planning process begins at the network-wide strategic planning level. At this level, a firm must determine how warehouse operations fit into its overall strategic plan, and in particular, what the mission is of the warehouses on its network. Figure 2

Figure 1: Hierarchical Supply Chain Planning Framework Corporate

Objectives Product/Market Mix

Manufacturing Plans

Strategic (2 yrs. + )

Capacities Facilities Locations Resources

Constraints

Aggregate Production / Distribution Planning Allocates capacity and resources to product lines

Tactical (12 to 24 months)

Assigns sales regions to DC’s and plants Constraints

Operations Scheduling Distribution Resource Planning (DRP)

Master Production scheduling Short run DC workload scheduling Transport scheduling

Short-term Scheduling (shop floor)


Operational (1 to 18 months)

provides a high-level overview of this hierarchical planning process that begins at the strategic level. A first step in the planning process consists of determining the mission of the overall warehouse network and the individual locations that will make up the network. Note that not all warehouses on a network will necessarily have the same mission or play the same role. The number of warehouse echelons to establish represents another common strategic network design question that heavily influences the mission of individual warehouses. For example, a firm must decide whether it will operate a single echelon network in which every warehouse will receive shipments of all products directly from all plants, or alternatively does the firm want to operate a multiechelon warehouse network where one or more first echelon, central warehouses receive products from plants and then redistribute some or all products to second echelon regional warehouses. Another important strategic decision concerns the question of whether a firm chooses to operate its own facilities or to outsource some or all of its warehouse operations to third-party providers. Finally, as Figure 2 illustrates, total network warehouse capacity requirements and the economics of scale trade-offs are two additional key determinants of the interrelated decisions on network design, facility design and warehouse technology selection. At the tactical level, a firm must concern itself with such planning activities as balancing the demand for warehousing capacity across its network and planning the most efficient and effective utilization of its capacity at each individual DC. Capacity planning at the individual DC level can involve determining the overall labor level and mix required to meet the projected demands over the planning horizon, the proper mix and use of available storage locations (e.g., type of racking where adjustable) and so on. In general, tactical warehouse planning focuses on the determination of how to best employ the existing network infrastructure (i.e., the existing warehouses and material handling equipment). Additionally, decisions to purchase relatively minor additional warehousing assets (e.g., incremental material handling equipment, racking, etc.) will occur in the tactical planning process. Major infrastructure issues that a firm cannot resolve at the tactical planning level (e.g., inadequate network capacity to meet forecast long-term warehouse throughput or storage requirements) must typically be fed back up to the strategic planning level for resolution. Thus, the efficacy of hierarchical warehouse planning and scheduling relies upon feedback loops, similar, for example, to the dependency of effective production planning on such mechanisms. At the operational level, a broad assortment of warehouse planning and scheduling activities take place on a regular basis. Figure 3 illustrates a sample of key decisions that operational schedulers must address. The scheduling of labor and short-term assignments of items to storage locations represent two of the major operational planning activities. Typically, it is the non-routine components of these activities (e.g., addressing temporary labor or storage requirements that significantly exceed capacity) that require the most critical attention. It is also typically the exceptions or non-routine requirements of operational planning and scheduling that planners must report or “feedback” to the tactical planning level. For example, when warehouse planners consistently find themselves having to schedule “unplanned” outside storage because of insufficient facility storage


77

Frameworks for Continuous Distribution...

capacity, they should send this information to the tactical level for resolution. Perhaps the overall warehouse network is out of balance and requires realignment because excess storage capacity exists at certain warehouses, while other warehouses face the opposite situation. Alternatively, perhaps this storage capacity issue at one warehouse is not an imbalance issue, but rather is occurring regularly across the network and requires a total network solution. This represents just one simple example of the types of feedback loops that must exist between the operational and tactical warehouse planning levels.

Supply Chain & Logistics

Figure 2: Hierarchical Warehouse Planning Network Design and Warehouse Location Overall Network Capacity Number of Echelons

Strategic

Facility Design and Technology Selection Scale Trade-Offs Constraints

Benefits of Frameworks – Summary

In this article, we have reviewed the general concept of using frameworks to facilitate the organization of planning and scheduling activities on a supply chain. We have noted that frameworks have broad applicabilty to most to all major functions and processes. To illustrate this generic applicability, we have examimed a framework for the ongoing maintenance and evaluation of warehouse operations. Now, in closing, we consider additional benefits of the framework approach. A firm that actively employs frameworks for its supply chain functions can rapidly evaluate whether potential new processes or decision support tools (e.g., scheduling software) can enhance the efficiency of its supply chain. Specifically, through its frameworks, a firm’s management understands how its current supply chain activities interface with each other at the operational, tactical and strategic level, as well across these levels. When considering a potential addition or revision to its supply chain, the firm’s decision makers evaluate how the proposed addition or revision fits into its existing framework. What are the new interfaces and communications that would be required? Does the proposed addition or revision really just duplicate a capability that already exists or does it truly improve a function or process? These represent the types of important questions to which well maintained frameworks facilitate insightful, broad-based answers. In the absence of frameworks, firms have a greater risk of making ineffective supply chain decisions. Further, maintaining good frameworks of each major function can accelerate the assimilation of new tools and methodologies

Aggregate Planning Capacity Balancing Across Network

Capacity Planning Within DC Sku/Item Location and Allocation

Tactical

Constraints

Operating Procedures and Policies

Operational Daily and Short Run Scheduling

into a function. Once managers decide to introduce a new methodology or decision support tool, the perspective provided by their frameworks of the supply chain functions allows them to best leverage the utility of the new capability. In summary, the practice of maintaining good frameworks of a firm’s major supply chain functions represents a valuable and potentially differentiating capability to assure an efficient and effective supply chain. References 1. Miller, T.C. (2002), Hierarchical Operations And Supply Chain Planning, Springer, Heidelberg, Germany. 2. Liberatore, M.J., and Miller, T.C. (1998), A Framework For Integrating Activity-Based Costing And The Balanced Scorecard Into The Logistics Strategy Development And Monitoring Process, Journal Of Business Logistics, 19 (2), 131-154.

Figure 3: Illustrative Operational Warehousing Decisions

Operational Planning Horizon

• What assignment of customer orders to the different types of pick operations in a warehouse will maximize operating efficiency? • How much space should be allocated for different product types and different activities? • What items should be diverted to temporary outside storage when storage space requirements exceed short term capacity? • How should individual jobs be scheduled in the warehouse?

Tan Miller, Regional Director of Distribution at Johnson & Johnson Group of Consumer Companies, is currently responsible for the operations of Johnson & Johnson’s U.S. Consumer Distribution Network. Previously, he has held similar responsibilities for Pfizer, and he has also held production and distribution positions with Mercer Management Consulting, Unisys and American Olean Tile Company. Tan has published four books and over 50 articles on supply chain and logistics operations and planning.

78



If you had a toothache, you wouldn’t try to ﬁx it yourself...

would you? You’d trust someone who already has the equipment, experience and expertise. At DDN, we understand there’s a lot to know about launching a successful life sciences company or product and how to manage the complicated process along the way. We think of every detail – including the things you probably haven’t even considered. DDN offers: • • • • • •

Product storage in secure warehouse facilities, including climate control Order receipt, processing and fulﬁllment Financial service (order to cash) Contract management, including Medicaid and chargebacks Sample and literature distribution Technical services and reporting

DDN Pharmaceutical Logistics is a full-service, third-party logistics provider serving the life sciences industry. We are the largest and longest-standing privately held provider in the industry.

For more information on how DDN can help your company save, call

8 8 8 .374.8 873 www.ddnnet.com