“Exploring the limitations of the Customer Order Decoupling Point concept”

“Exploring the limitations of the

Customer Order Decoupling Point

concept”

By: O.J-C. Octobre Student nb: S1587145

University: RijksUniversiteit Groningen

Faculty of Managment & Organization

Specialization: Operation & Supply Chain Management

Organization: Litterature Thesis

Supervisors: Dr J. de Vries

Dr J.T. Van der Vaart

--- Rijksuniversiteit Groningen

- 1 -

Preface

The following report is the final thesis that marks the end of my study in Business Administration at the RijksUniversiteit in Groningen.

As a foreigner student I would like to thank Groningen University for allowing me to follow my studies in the Netherlands and the faculty of Management and Organization for accepting me in their program. I would especially thank Martin Land my program coordinator and teacher for his support and his advices during all this year.

I would also like to thank Jan de Vries and Taco van der Vaart my thesis supervisors, who gave me useful advices and feedback that helped me a lot to achieve this research.

I would like to say special thanks to Guillaume Chouteau that was my contact at Caterpillar in France, who dedicated a lot of time in order to give me information. I am also very thankful to Dr J. Wijngaard and Dr J.C Wortmann for their participation to my questionnaire.

--- Rijksuniversiteit Groningen

- 2 -

Executive summary

--- Rijksuniversiteit Groningen

- 3 -

Index

1.3. MAIN RESEARCH QUESTION...10

1.4. SUB QUESTIONS...10

1.5. RESEARCH OBJECTIVE...11

1.6. RESEARCH DESIGN...12

1.7. DEFINITION OF KEY CONCEPTS...13

1.8. STRUCTURE OF THE RESEARCH...15

2. ...15

CHAPTER 2. GENERAL LITERATURE REVIEW ABOUT THE CODP...16

2.1. WHAT IS A CODP? ...16

2.2. THE ROOTS OF THE CODP ...17

2.2.1. The birth of the CODP concept...17

2.2.2. The Dutch stream ...17

2.2.3. The American stream ...19

2.2.4. The recent stream...21

2.3. BASIC CONCEPT OF THE CODP ...23

2.3.1. The bases of the CODP...23

2.3.1.1. Time competition ...23

2.3.1.2. Mass customization...24

2.3.2. The benefits of the CODP...25

2.3.2.1. Cost reduction...25

2.3.2.1.1. Inventory reduction ...25

2.3.2.1.2. Other cost reductions ...26

2.3.2.2. Flexibility enhancement ...27

--- Rijksuniversiteit Groningen

- 4 -

2.4. WHAT ARE THE FACTORS THAT INFLUENCE THE CODP...29

2.4.1. Effects of back and forward shifting...29

2.4.2. Product related factors...30

2.4.3. Market related factors...31

2.4.4. Process related factors ...32

2.5. MULTIPLICITY OF THE CODP...36

2.6. EXTENSIONS OF THE CODP CONCEPT...37

2.6.1. Other decoupling points ...37

2.6.1.1. Information decoupling point ...37

2.6.1.2. Engineering decoupling point ...38

2.6.2. The postponement concept ...40

2.6.3. The modularization concept ...41

2.6.4. The leagility concepts...42

2.6.5. The consequence of these theories for the CODP ...43

2.7. Conclusion ...44

3. P...45

CHAPTER 3. DISCUSSION ABOUT THE LIMITATIONS OF THE CODP...46

3.1. PROBLEM WITH DETERMINING THE POSITION OF THE CODP ...46

3.1.1. Machine setup and batch size...47

3.1.2. Bottleneck...48

3.2. PROBLEM WITH SHIFTING THE CODP ...49

3.2.1. Adaptations of the systems...49

3.2.2. Form postponement demand excess capacity ...50

3.2.3. implication of modularization ...51

3.3. PROBLEM WITH DECOUPLING THE FLOWS...51

3.3.1. Impact of the multiplicity of CODPs...51

3.3.2. Impact on the production control and planning process ...52

--- Rijksuniversiteit Groningen

- 5 -

CHAPTER 4. QUESTIONNAIRE...53

4.1. OUTCOME OF THE QUESTIONNAIRE...55

4.1.1. Question 1 : Positioning of the CODP, barriers...55

4.1.2. Question 2 & 3 : Shifting of the CODP, production planning and control process ...55

4.1.3. Question 4 : Shifting of the CODP, modular manufacturing ...56

4.1.4. Question 5 : Decoupling the flows, the case of multiple CODP ...56

4.1.5. Question 6 : Positioning of the CODP, impact on the processes...57

4.2. Discussion about the questionnaire ...57

5. Q ...57

CHAPTER 5. CASE STUDY...58

5.1. PRESENTATION OF THE FIRM...58

5.1.1. General presentation ...58

5.1.2. Product produced by Caterpillar ...59

5.1.3. Kind of production ...60

5.1.4. Supply Chain at Caterpillar...62

5.1.5. The order delivery process...62

5.1.6. Production planning and control ...63

5.2. CODP AT CATERPILLAR...66

5.2.1. Position of the CODP ...66

5.2.2. Reasons for this positioning ...66

5.3. LIMITATIONS OF THE CODP AT CATERPILLAR. ...69

6. 5 ...71

CHAPTER 6. CONCLUSION...71

6.1. HYPOTHESIS AND MAIN RESULTS...72

6.2. SHORTCOMINGS OF THE RESEARCH...74

6.3. SUGGESTION FOR FURTHER RESEARCH...74

7. TABLE OF FIGURES ...75

TABLES...75

FIGURES...75

--- Rijksuniversiteit Groningen

- 6 -

Chapter 1. Introduction

1.1. Background

The Customer Order Decoupling Point (CODP) also called the Order Penetration Point (OPP); is the position in the value chain where the information about the customer order gets known. This point make a distinction between two parts in the value chain: the part downstream the CODP that is based on known orders data and the part upstream the CODP that is based on forecasts. The Customer Order Decoupling Point is always a stock point. This can be a stock of raw materials, components or already assembled products. This stock point represents the last stock point of the value chain. The replenishment of this stock point by definition does not depend on customer orders information but it depends on forecast of the demand. All stock points upstream the CODP are based on forecast. Once an order arrives all the products upstream the CODP are already available because produced, purchased or assembled on stock. The usage of this stock point is then determined by the information of the customer order. The received orders will help to determine the number of product to be processed downstream the CODP. Downstream the CODP the products are processed with little need of inventory. The right amount of product to produce is known so production is planned according to this number.

Figure 1. The CODP concept

CODP

Value Chain

_{Value Chain}

Order based operations Order Fulfilment process Raw

materials

Arrival of an order

Forecast based operations Make to forecast process

--- Rijksuniversiteit Groningen

- 7 -

In the value chain, each time an item needs to be purchased, produced or assembled is called a planning point. “An item is a collective term for any good that can or must be identified or handled in the logistics of distribution, production, procurement or recycling/disposal” (Schönsleben, 2000) “A planning point is a manufacturing resource such as a work centre that can be regarded as one entity from a production and planning point of view” (Olhager, 2003). In fact a planning point represents a point in the value chain where a decision needs to be made about quantities. This decision is made on a certain frequency: once per month, per week or per day. A planning point can be either order driven or forecast driven. The position of each planning point can be calculated with lead times. The time at which the item is needed plus the replenishment lead time of this product represents the position where decision must be made. If at this time the order information is known then the inventory is not necessary and decision is order driven. If at this time the order information is not known then forecast is needed to determine the level of stock and decision is forecast driven. CODP represent the point in the value chain that separates these two modes of decision making.

Separation between order and forecast driven decision making is present in most of firms. However shifting this point in order to reach the optimum position according to the strategy is not always easy. Certain characteristics of the firm may create constraints for shifting the CODP or for making a clear separation between the two parts.

Within the concept of CODP we can find a relationship with a lot of topics that are of high importance for supply chain management. This explains why CODP is strategic for a firm. First of all the CODP is closely related to stock management and how to reduce the inventory cost effectively. It gives an answer by explaining that forecast based stock must be reduced to the smallest part as possible. To reduce inventory it is very important then to shift the CODP upstream. The reason is that downstream the CODP less stock is needed.

--- Rijksuniversiteit Groningen

- 8 -

adding chain depending on product, process and market factors. A good placement of the CODP will help to improve the cost/service trade-off because an operational strategy will be put in place according to the customer itself. What are these benefits? What are the factors that influence the position of the CODP? These questions have to be answered in order to make the usage of this concept successful. One very important point is the Cost-service trade-off. As we stated before shifting CODP upstream may lead to decrease inventory costs but this very often comes with the decrease of the service level (longer delivery lead time). One the other hand shifting the CODP more close to the customer (downstream) would lead to increase of service but also increase of costs. To solve this problem it is necessary to overcome the cost-service trade-off by for example reducing lead time.

There exist today a lot of other concepts that are related to the CODP concept. This is the case for the postponement concept, the concept of leagility or the modularization concept. All these new concepts are extensions of the decoupling point concept. The main idea of these concepts is to give an answer to the question: how to manage to shift the CODP effectively along the value chain? This is because the difficulty in the application of the CODP is to be able to shift it in order to reach the most suitable position. If these concepts are today well known; only little case studies about the application of the CODP have been done. Especially in the light of the recent publication about leagility or postponement it is interesting to have a look at the limitations that appears when trying to position the CODP in a specific place.

--- Rijksuniversiteit Groningen

- 9 -

1.2. Problem statement

The concept of the Customer Order Decoupling Point is a rather old concept that dates back from the early 80’s. Several authors at this date formalized the fact that there exist two kinds of decision making in a value adding process. The separation between these two parts is made by the introduction of customer order information in the value chain. The more the information arrives at and early stage of the value chain the more decision making is order oriented and the fewer inventories is needed. The CODP role in this context was first to determine where this information should be introduced in the value chain in order to fulfil customer requirement (requirement of a specific delivery lead time) for the cheapest cost. Nowadays this concept is broadly recognized by firms because it is a strategic issue that relates to the cost-service trade-off.

As we can see through the literature the most difficult part for a firm is to manage to position its CODP according to its strategy. Sometimes the required position of the CODP is not possible to reach because of long lead times or cost issues, sometimes the process of the firm are not adapted to the strategy. The factors that influence the positioning of the CODP need to be well identified and controlled in order to be able to position the CODP at the most suitable position. There are several ways to shift the CODP upstream or downstream and they all relate to the factors influencing the position of the CODP. We are going to focus especially on two possible ways: postponement and modularization.

--- Rijksuniversiteit Groningen

- 10 -

Author

Study

Findings

Haan, Yamamoto and Lovink (2001)

Production

planning in Japan • CODP is crucial to define the production planning process of one firm Van Donk (2001) Decoupling point in

the food processing industry

• CODP concept can be transformed into an applicable decision aid for managers. • Finding a balance between the different

factors that affect CODP placement is the most difficult.

• Organizational and cultural barriers need to be overcome to implement CODP

Kisperska-Moron (2003) Inventory decision in Polish manufacturing companies

• Inventory decision through CODP implementation is not integrated in the majority of Polish companies. It result in difficulties to plan and realize strategies. Meijboom (1999) Production to order

and international operations in the clothing industry

• The CODP in each product market-combination is highly important in international environment. All different combination must be very well coordinated to integrate the whole process of the firm

Table1. Summary of case studies about CODP implementation

1.3. Main research question

What are the limitations of the CODP concept?

1.4. Sub questions

--- Rijksuniversiteit Groningen

- 11 -

application of the CODP for a firm that are related to the positioning of the CODP at the best suitable position. This results in the following sub questions:

• What are the roots of the CODP • What is a CODP?

• What are the benefits of CODP?

• Which factors influence the position of the CODP? • How to shift the CODP along the value chain?

1.5. Research objective

--- Rijksuniversiteit Groningen

- 12 -

1.6. Research design

The research is divided in three main phases that are general literature review about CODP, expert interview and case study.

Figure.2 Main research phases of the thesis

This research is a qualitative exploratory research that aim is to find out what are the limitations of the CODP concept. The data collected will come from expert interview, case study and secondary literature review.

During Phase I a large literature review will be done in order to have a complete overview of the CODP concept. Through review of the most important literature about CODP we will focus on answering the sub questions of the research. These questions will help us to point out the main issues related to CODP.

At the end of phase I we will have a clear view of all the different factors that influence the ability of a firm to shift the CODP. As we already stated the ability to move the CODP along the value chain is essential if a firm wants to successfully use the CODP concept. This will then lead us to examine which difficulties a firm may encounter to shift the CODP.

In phase two we will focus on analysing the answer of the questionnaire. Operation management experts were chosen for this questionnaire because the CODP is an operation management topic. Such topics as Inventory management, order processing or production planning and control are linked to CODP and relate to

Literature

review

Expert

interview

Case study

--- Rijksuniversiteit Groningen

- 13 -

operation management. Questions about the possible limitations of the CODP concept will be asked. Topics like difficulties in shifting the CODP, separating the decision making structure and introducing the customer information at a precise position of the value chain will be tackled.

Phase three will aim at gathering real data about the practice of the CODP in a firm. A short case study of the firm Caterpillar in France will be done. This study will focus on the specific limitations and problems highlighted in phase 2 and 1. After presenting the firm (product, market and production process) we will find out which is the position of the CODP in the firm. Finding the position will lead to answer to several questions. Why is the CODP at this position? Is it relevant according to the strategy of the firm? Which are the consequences of this positioning? With the help of these data we will finally discuss the different limitations of the CODP concept and we will give conclusion and suggestions for further research.

1.7. Definition of key concepts

• CODP : Customer Order Decoupling Point

Several definition of the CODP exists. Among them the most used is the one of Hoekstra & Romme (1984): The decoupling point is the point that indicates how deeply the customer order penetrates into the goods flow.

• Value Chain

--- Rijksuniversiteit Groningen

- 14 -

CODP influences in fact more than the goods flow, it influences the whole value chain (Giesberts & Van der Tang, 1992)

• Mass customization

Mass Customization refers to a process of production of goods and services tailored to suit the needs of customers in a mass market (Selladurain, 2003)

• Production control and planning

According to Schönsleben (2000) a production control and planning system refers to administrative and planning logistics. Administrative logistics refers to sales order processing with regard to document, movement of goods or inventory. Planning logistics is related to issues such as when, how and in what quantity will goods be produced or procured. In our paper we will focus on the following administrative and planning logistics tasks: forecasting, order processing, distribution, project management, taking back, resource planning and inventory (Schönsleben, 2000).

• Planning point

“A planning point is a manufacturing resource such as a work centre that can be regarded as one entity from a production and planning point of view” (Olhager, 2003)

• Stock point

A stock point is a point in the value chain where material is stocked. It is linked with a physical stock pile. Generally a stock point is the result of a planning point we can link these two points like this:

Figure 3. Stock point and Planning point

Planning point Stock point

PRODUCTION PROCESS

Lead time of the production process

--- Rijksuniversiteit Groningen

- 15 -

1.8. Structure of the research

This research is going to be presented in six chapters

• Chapter one will be dedicated to the introduction of the research. Problem statement and research question as well as sub question will be presented.

• Chapter two will consist of a literature review about the CODP. The aim will be to define clearly what a CODP is. Benefits of the CODP and factor influencing the position of the CODP will be presented. Moreover a review of the most recent extension of the concept will be done. This chapter will aim at presenting a complete overview of the Customer Order Decoupling Point concept.

• Chapter three will discuss the limitations of the CODP concept that could be found in the literature.

• Chapter four will present the questionnaire and its outcome. Analysis of the answers will permit to find out new limitations and to illustrate the limitations already found in the literature.

• Chapter five will present the case study and its outcome. The case study was done at Caterpillar in Grenoble (France). The objective is to find out the position of the CODP in the firm. We will try to understand what were the choices leading to this position and what are the consequences for the firm. The data obtained will help us to illustrate the previous findings about the limitations.

• Chapter six will present the conclusion and give an answer to the research question. We will present our main findings and give some suggestions for further research.

--- Rijksuniversiteit Groningen

- 16 -

Chapter 2. General literature review about the

CODP

2.1. What is a CODP?

--- Rijksuniversiteit Groningen

- 17 -

2.2. The roots of the CODP

2.2.1. The birth of the CODP concept

The COPD concept as we can find it today was first mentioned by Shigeo Shingo in his book named “study of Toyota production system from industrial engineering viewpoint” in 1981. For the first time in this book it was mention of the P:D ratio. This ratio was then explained more deeply by Hal Mather (1988). Shingo is often cited in the literature about lean manufacturing. Lean manufacturing is in fact an extension of JIT manufacturing that was used at this time. Shingo explains that it is important to use JIT manufacturing to decrease costs but it is also important to do it with no loss of flexibility. P:D ratio is a tool that will help to determine where to use JIT manufacturing in order to keep flexibility in the value chain. The P:D ratio is in fact a tool to determine where to position the CODP. The JIT part represents the forecast driven part of the value chain where production must be done at the cheapest cost. The order driven part is linked with a more flexible production system. (Stalk & Hout, 1990).

2.2.2. The Dutch stream

--- Rijksuniversiteit Groningen

- 18 -

--- Rijksuniversiteit Groningen

- 19 -

Figure 4. Five main logistics structure (Towill, 2000)

These five logistic structures are a reference still today in all the literature. Moreover in their book, Hoekstra and Romme give some insight about how to find the best place for the CODP and the cost-service trade-off is at stake. They explain that the product but also the market itself and the combination of product/market must be taken into consideration while choosing the kind of logistic structure to be used. Delivery lead time required by the market is the first factor to take into consideration: the more the delivery lead time is short the more the chosen CODP must be close to the customer. The second factor is the price of stock holding. The more the cost of stock holding will be expensive the more the chosen CODP must be close to the supplier. Of course other factors enter into consideration when choosing a CODP and we will discuss them later in this paper.

2.2.3. The American stream

--- Rijksuniversiteit Groningen

- 20 -

uses the production lead time and compares it to the order fulfilment lead time that the firm is able to propose. The order fulfilment process is influenced mainly by the need of the customer. The willingness of the customer to wait for an item will determine the length of the order fulfilment process. The characteristic of the firm will determine what the firm is able to do according to the need of customers. Even if the customer desires a short lead time the firm may not be able to answer to it because of its processes. According to the ratio of these two lead times Mather can then determine where the last stock point should be placed in order to be able to fulfil customer requirement. After this stock point the firm is able to produce the finished product within the required delivery lead time. If the firm starts producing before this stock point it will not be able to finish the product within the expected lead time. This stock point will represent the earliest point where the order can be taken into consideration in order to respect the lead time expected by the customer. This stock point represents in fact the last point of the value chain where decision making is forecast oriented. It is generally assumed to be the point of incoming of customer order.

--- Rijksuniversiteit Groningen

- 21 -

Figure 5. The P:D ratio concept

The usage of the CODP is then very useful to keep a good ratio of costs comparing to the service level needed. Especially because this ratio takes into consideration the expected service level (delivery lead time) The P:D concept can of course be related to the framework of Hoekstra & Romme as we end up with the same logistics structure. However while Hoekstra & Romme focused more on the logistic structure, Mather gives a more detailed explanation of the forces that influence the positioning of the CODP. This aspect will be explained later.

It is important to notice that in this figure Mather only consider the production process. The purchasing and delivery process are also part of the value adding process and they must also be included in the calculation (Christopher, 1998) In this case we compare the delivery lead time with the sum of purchasing, production and delivery lead time.

2.2.4. The recent stream

--- Rijksuniversiteit Groningen

- 22 -

multiple or other kind of decoupling points: demand mediating decoupling point (Olhager, 2003) or engineering decoupling point (Wikner and Rudberg, 2004). Other authors discussed the impact of the CODP in different situations: Impact of the CODP for information system and production control (Giesberts & Van der Tang, 1992) or the usage of CODP in the food industry (Van Donk, 2000) among others. Finally a lot of new theories emerged using the CODP concept as a basis: postponement (Yang & Burns, 1999; Feitzinger & Lee, 1997), leagility (Naylor & Berry, 1999) and modularization (Gu & sosale, 1999).

Figure 6. Chronological history of the CODP 1981 : Shingo

1984 : Hoekstra & Romme

1988 : Mather

1992 : recent publications

1997 : Extensions of the concept

--- Rijksuniversiteit Groningen

- 23 -

2.3. Basic concept of the CODP

2.3.1. The bases of the CODP

As we could see in the history the CODP has its roots in different domains. On the one hand CODP is an extension of the JIT concept that aims to enhance flexibility and on the other hand it is a way to reduce inventory costs. The general definition of the CODP is that it is the point that separates the forecast driven operations from the order driven operations. This seems to be a simple concept but there are much more implications of this concept for the overall strategy of one firm. The division of the value chain into two distinct parts creates space to improve both parts in different ways. Moreover the use of CODP leads to reshape the logistic structure of one firm. CODP is in fact a dynamic concept that enables but also influence the design of a supply chain. Recent papers summarized more clearly the basic reason of usage of CODP

2.3.1.1. Time competition

Because of the recent changes in the economical environment, competition on the time element (time for producing, time for innovating, time for delivering) is becoming a main point for manufacturing firms. Increase of competition, globalization of markets and the shortening of product life cycle make the delay to provide the market with one product a very important point if one wants to keep its market share (Stalk 1988). Value added by manufacturing is lower than in the past and retailers are the dominant player in most supply chains (Sharman, 1981), thus the manufacturers need to provide the market with new competitive advantages. Improve of quality and service is one of these advantages. Innovating faster or delivering faster than the competitors is a key competitive advantage.

--- Rijksuniversiteit Groningen

- 24 -

ten years the time based competition has been a very important element influencing the strategy of a lot of firms (Stalk & Hout, 1990). The CODP concept has been widely applied in this context in order to redesign the value chain with the aim of decreasing costs to keep being competitive. The CODP is then directly linked to the competitiveness of one firm and permit to redesign the firm in order to answer better to customer’s specific needs. The more speed is an order winner element the more the CODP has to be located downstream the value chain (Towill, 2005) in order to keep competitiveness. This means that products are produced to stock in order to have them available for the customer in a very short time.

2.3.1.2. Mass customization

Another reason that leads to shift the CODP more upstream in firms is the increasing need of customization of the products. More than the time based competition the markets are today characterized by an increasingly widening of the demand of the customer. A standardized product is not anymore a successful strategy today. It is replaced by mass customization (Towill, 2005). In this case the CODP enhances the flexibility downstream the value chain. The forces that influence the position of the CODP and shape the value chain are then productivity against flexibility trade-off (Rudberg & Wikner, 2004). Upstream the CODP the processes are characterized by productivity issues: economies of scale, large batches. While downstream the CODP the processes are more dedicated to flexibility: small batches, large diversity.

--- Rijksuniversiteit Groningen

- 25 -

2.3.2. The benefits of the CODP

As we have just seen, the CODP concept has been applied into companies mainly because of the rise of time competition and the need for mass customization. Most of the firms had to reshape their logistic structure during the last decades because of the changes of the market environment. To shift the CODP leads to reshaping the logistic structure in order to respond to the demand of the market more efficiently. What we are going to speak about in this part is the benefits that can be obtained through a good fit of the position of the CODP with the strategy (the benefits of a right position of the CODP). CODP may be placed upstream or downstream it always leads to positive impact for the firm when it is suitable. These benefits are however not always clear so we are going to explain them.

2.3.2.1. Cost reduction

2.3.2.1.1. Inventory reduction

--- Rijksuniversiteit Groningen

- 26 -

reduction downstream the CODP. Moreover the cost of stock holding increases with the value added to the product so it is less expansive to stock raw materials than finished products (Hoekstra and Romme, 1984). If the overall level of stock is not decreasing because the stock point are just moved from one step to another, there is still a reduction of inventory cost because of the lower cost for holding the same amount of stock.

DP1 = make to stock

DP5 = make and buy to order

Figure 7. The impact of the CODP on the stock investments (Hoekstra & Romme, 1984)

2.3.2.1.2. Other cost reductions

In the mean time as inventory cost can be reduced we can also notice several other cost reduction thanks to the CODP. First the transportation costs are reduced because of the decrease of uncertainty. There is less to transport because no inventory and the load of trucks can be scheduled more efficiently with accurate data. More

Business risk in ensuring delivery performance

Business risk in stock investments

Business risk In resources

--- Rijksuniversiteit Groningen

- 27 -

generally the reduction of uncertainty always tends to reduce costs (Mason-Jones & Towill, 1998) thus upstream the CODP the costs tends to be reduced in general.

One the other hand the more the process will be oriented to order the more cost will tend to increase because of short lot size and smaller batches (Goyal, 1993). Setup costs and transportation costs are increasing when the lot sizes are smaller. That is the reason why the CODP concept cannot be introduced in the same way for all firms because all have different opposing forces.

One very important factor that leads to cost reduction is also the capability of productivity enhancement enabled by the CODP. The appliance of lean thinking is very often advised in the upstream part of the value chain (Naylor, Naim & Berry, 1999). Indeed, upstream the CODP, the logistic structure can be focused on lead time, quality and cost in a so-called push process.

What the CODP makes possible is the ability to overcome trade off between cost and service. By focusing on reducing cost having in mind the level of service required by the customer (the expected delivery lead time) the firm is able to satisfy its customer at the cheapest cost. The precise placement of the CODP and the effective reengineering of both side of the value chain is a really useful tool to reduce cost by keeping an acceptable level of service. Overcoming this trade-off is possible because the strategy employed to reduce costs has for referent the expected lead time. So the reduction of cost does not lead to a reduction of the service level.

2.3.2.2. Flexibility enhancement

--- Rijksuniversiteit Groningen

- 28 -

As we stated before recent theories are trying to give some insight in order to overcome the traditional trade-off between cost and service that appears generally. In term of value chain we can speak about the productivity-flexibility trade-off (Wikner & Rudberg, 2004) This trade-off represent the constant questioning between having a flexible system that answer to customer demand effectively or having a productive system that is able to produce with cheap costs. Both system have as a goal to gain a competitive advantage but recently the researcher have highlighted the fact the you cannot just choose one strategy and a mix between agility and productivity must be achieved, that is the basis of the concept of leagility that we will discuss later (Naylor Naim & Berry, 1999) As well as with traditional trade-off curve the position of the CODP will result in more or less flexibility.

Overcoming the trade-off is today the main problem and this result in trying to move the CODP downstream or upstream while keeping costs at a good level. Technology especially is helping to improve this aspect (Hull, 2005). CODP and focus on flexibility is a first step to acquire a so called efficient mass customization. It enables to meet the moving demand and to satisfy customer with specific needs.

2.3.2.3. A strategic fit between the resources and the market

--- Rijksuniversiteit Groningen

- 29 -

resources of the firm in order to reach the expected delivery lead time. The CODP is thus a very good strategic tool to reach the fit between resources and the market.

It is possible that a firm will need to change its CODP over time. This will depend on the characteristics of the product-market that are also called the order winners (Towill, 2005). An order winner is a characteristic of one product that will convince a customer to buy a product. If the order winner is a short delivery lead time it will lead a firm to shift the CODP downstream. If the order winner is a high degree of customization of the products it will lead a firm to shift the CODP upstream. This shows that the CODP is very strategic.

2.4. What are the factors that influence the CODP

As mentioned before a good position of the CODP is a key to create an effective logistic structure. However the CODP is influenced by a lot of factors that must all be taken into consideration. At first the CODP is depending on the product/market. The demand, the characteristic of the product will have a direct impact on the possible placement of the CODP. But also the process of the firm and the technology is a constraint that will impact the placement of the CODP (Olhager, 2003). Thus the final decision of the placement of CODP is in fact a very difficult one and all the factors must be taken into consideration

2.4.1. Effects of back and forward shifting

--- Rijksuniversiteit Groningen

- 30 -

While the P:D ratio and the product/market characteristics help to determine the optimal position of the CODP ( Mather, 1984; Hoekstra & Romme, 1988) it is not always possible to have the CODP at the more suitable position because of to the existing characteristics of the firm. For example a firm that produces to stock realizes that it has to produce to order to be more competitive. To realize this change the firm will have to move the CODP upstream. In order to shift the CODP it is necessary to make modifications on the factors that influence the position of the CODP. Reducing the delivery lead time leads for example to upstream shifting of the CODP. Several possibilities exist to reduce the delivery lead time: increasing the efficiency of the delivery process with the help of automation is one of them. (Olhager, 2003). Forward or backward shifting of the CODP is very useful to reach the optimized position for the CODP. What is the main point of the CODP concept in practice is to know how to shift it and to manage to shift it efficiently.

2.4.2. Product related factors

First group of factors that affect the CODP position are the product related factors. The degree of modularity of the product and the customization opportunities are to be taken into consideration. It is unconceivable to make on stock if there is a large variety of product because it will lead to high level of stocks. In the same way a product with a modular structure will prefer to have an Assemble to Order configuration (Olhager, 2003). The level of customization of the product is a force that leads to shift the CODP upstream, but to be moved upstream the product must also have a corresponding material profile. Each different variety of the product must have some common parts or be easily modularized (Van Hoek, 1995). The CODP is particularly important for product with an X material profile (Olhager, 2003), downstream the CODP product have a convergent structure while upstream they have a divergent structure.

--- Rijksuniversiteit Groningen

- 31 -

it can get damaged) are the main characteristics of the product that will increase the inventory handling cost (Stock & Lambert, 2000). Some of these factors will result in upstream moving of the CODP while other will tend to lead the CODP downstream. To these we can add the perishability of the product (in the food industry for example) that has a direct effect on the CODP (Van Donk, 2001).

Modularity of the product is another factor that influences the position of the CODP. In fact it is a downstream force because the more one product has a modular structure the more components and sub assemblies can be produced in advance before assembling the final product. So finally to information about order that will lead to the differentiation of the product are not needed in advance. Most of the components that will be combined to create the final product are produced to forecast. The reason why they are produced to forecast is that they are “general” product. There is no risk of obsolescence of theses components that are also called module because the demand is not a specific demand. Moreover there are a relative low number of modules so the inventory is not too high. However it is important to keep in mind that not all the product is modular and there will still be a last step of production that will lead to differentiation of the product and for this step the CODP is needed.

2.4.3. Market related factors

--- Rijksuniversiteit Groningen

- 32 -

factors appear to be order winner (Towill, 2005) then it is wise to place the CODP in order to fulfil market requirement on this characteristic. Olhager (2003) gave different name for these factors: the product range and product customization requirement, the product demand volatility and volume and the frequency and seasonability of demand. These factors are closely related to the previous one but they are more detailed. For example a high volume of demand will move the CODP more downstream because of less risk of stock obsolescence. In the same way a seasonal demand will ask for more production to stock during peak period and thus the CODP.

Another important factor related to the market is the distance from the stock point. If a firm is delivering a product that needs to be transported over a long distance, it will increase the delivery lead time and thus it will have an impact on the position of the CODP. (Stock & Lambert, 2000)

2.4.4. Process related factors

The actual production lead time is the first factor that will influence position of the CODP because of its implication on the P:D ratio. As well as the required delivery lead time is important for the P:D ratio from a market point view, the production lead time is important from an inside point of view. In general the production lead time is a very important factor that determines the position of the CODP. Thus the existence of an efficient production process will have a high impact on the position of the CODP because the more the production process is efficient the more the lead time tends to be short (Mather, 1988). The reason is that when all steps of production are controlled and planned effectively it is easier to avoid queues, shortage of component and lack of capacity that will increase the lead time.

--- Rijksuniversiteit Groningen

- 33 -

a job shop, a planning point could correspond to a machine or a group of machines with similar capabilities while in repetitive production this would correspond to a line producing a single product family. In order to use the planning point it is necessary to have information about orders going through this point and this is where the CODP is used. The CODP is always positioned on a planning point because there is no need of information about orders between two planning point. Planning points are often linked to physical stock point but they are different. Each planning point is linked to a machine or a group of machine that processes products. Stock pile or queues may exist in front of these machines.

Also the value added between each stock point has to be observed in order to define the cost of stock holding at each step (Hoekstra & Romme, 1984). Price of stock holding is also a characteristic that is related to the processes, the more the process of stock holding is efficient the cheaper the cost of stock (Stock & Lambert, ). Position of the bottleneck inside the production process is another factor that influences CODP because of the need of full capacity utilization of the bottleneck some operations will have to be made to stock even if the CODP tends to favourize a make to order situation. So the existence of bottleneck will give an extra constraint to the placement of the CODP. Also the overall control of the processes is a force that is a constraint to flexibility thus that moves the CODP downstream (Hoekstra & Romme, 1984). The existing overall flexibility of the production process is in fact very important because of the higher customization that it enables. The higher the available customization the more upstream goes the CODP (Olhager, 2003).

Set up and batch size have also a very important role because the processes are constrained to these characteristics. Large batches or long setup times are often linked with a make to stock structure that leads to position the CODP more downstream. Also cyclic manufacturing is a characteristic that is best combined with a make to stock structure.

--- Rijksuniversiteit Groningen

- 34 -

the use of internet has been proven to lead to cost reduction and lead time reduction (Huank & Mak, 2001). An efficient order processing process will have an impact on the delivery lead time of one firm. Reduction of the lead time of the order processing will help the upstream movement of the CODP.

---

Rijksuniversiteit Groningen

- 35 -

Figure 8. Summary of the main forces affecting CODP position found in the literature Upstream forces

• Product factors :

o _{High value product}

o _{Complex product or product structure} • Market factors:

o _{Irregular demand, lack of}

predictability. Volatility, volume and frequency of demand

o Level of customization required • Process factors:

o _{Stock cost consideration (risk of} obsolescence).

o Flexible process

o _{Fast order processing process} o _{High usage of technology} o Decouplability of manufacturing

process, number of planning points

Downstream forces • Product factors:

o _{Modularity of the products} o Perishability

• Market factors:

o _{Delivery requirement (short lead} time expected).

o Closeness of the market o _Competitors

• Process factors :

o _{Long lead time,} o Long set-up time.

o _{Bad control and planning process,} dependability of processes

CODP

--- Rijksuniversiteit Groningen

- 36 -

2.5. Multiplicity of the CODP

The multiplicity of the CODP is a very important characteristic of CODP. Generally several CODP exists if there is product, process of market hybridism.

Product hybridism is characterized by the variety of the product and mainly by the variety in the product structure. Different product families with different product lines but still with common components introduce complexity inside the processes of one firm. It is not always possible to have only one CODP because the different products have different constraints and different production lead time.

Market hybridism is characterized by different markets for the same product. Each market has its own constraints (transportation cost, customer needs, order winner) Thus they require different treatment and again in some cases it is important to have several CODP.

The final case of hybridism is the process/production hybridism. The existence of different production situation is sometimes compulsory in one firm. A firm producing to order with large batches will be confronted to the problem of having to produce to stock if batches are not full of orders. In this case the CODP is not at the same position and two CODP can be used.

--- Rijksuniversiteit Groningen

- 37 -

2.6. Extensions of the CODP concept

If the literature about the basic concept of CODP is relatively scarce there is plenty of other theories that refers to CODP as a basis for their principle. In this part we are going first to have a look at theories that are employing other kind of Decoupling Point and then we will have a short overview of the theories about postponement, leagility and modularization.

2.6.1. Other decoupling points

Several other decoupling points have been identified by authors (Rudberg & Wikner, 2005; Mason-Jones & Towill, 1999; Olhager Seldin and Wikner, 2006) in the last few years.

2.6.1.1. Information decoupling point

--- Rijksuniversiteit Groningen

- 38 -

information is not distorted and is positioned as upstream as possible. The position of the information decoupling point does not depend on product type, degree of customization of supply chain strategy adopted. Information decoupling point only refers to information flow. The most important about the information decoupling point is that it refers to “pure” information that is not distorted. It is a tool to have a view on the information in order to differentiate accurate and inaccurate data. In the CODP literature there is no distinction between these two points moreover the information used at the CODP may already be distorted.

The authors generally states that generally the information decoupling point and the material decoupling point are in the same position. But Mason-Jones & Towill (1999) have a different point of view: this positioning is very wasteful and limits the effectiveness of the high value resource of undistorted order information available on dynamics of the supply chain (Mason-Jones & Towill, 1999). Generally the material decoupling point is placed as close as possible to the customer to facilitate the postponement strategies. This is in opposition with the fact that information decoupling point should be positioned as close as possible to the supplier. That’s why it is necessary according to authors to separate these two decoupling points otherwise relevant data are not used. In fact we will on later that dividing the CODP in two distinct decoupling points may be a solution to answer to the limitation that CODP may encounter.

However we need to put a limitation to this concept because it only refers to organizations with high level of customization. When the level of customization is high it is preferred to have little inventory and thus the material decoupling point is positioned downstream.

2.6.1.2. Engineering decoupling point

--- Rijksuniversiteit Groningen

- 39 -

engineering dimension. The result is an engineering decoupling point and a production decoupling points that are both in a distinct part of the value chain. While the production decoupling point deals with material the engineering decoupling point is used to manage the capacity of design and engineering of one firm. Rudberg & Wikner (2005) determined the same kind of structures as Hoekstra & Romme (1984) but on an engineering point of view. They explain for example that engineering to order is possible but also engineering to stock to a certain extent. The engineering decoupling point is in fact a normal customer order decoupling point that corresponds to a structure of engineer to order that we can find in the CODP literature. The importance of this decoupling point is that product can be engineered to stock.

While the customer order decoupling point generally refers to a material decoupling point this decoupling point bring the involvement of the customer more upstream of the process. Before producing any material the customer may be involved in engineering the product.

Figure 9. Engineering decoupling point

According to figure 9 the engineering decoupling point and the material decoupling point may be placed anywhere on the value chain with respect that they stay in their respective process. This results in several production and engineering

Material decoupling point engineering

decoupling point

Raw materials Finished product

--- Rijksuniversiteit Groningen

- 40 -

couples. A product can be Engineered to Stock and Manufactured to Order. Then the engineering decoupling point is downstream the engineering process while the material decoupling point is more upstream the manufacturing process. A product can also be Engineered and Manufactured to Order or Engineered and Manufactured to Stock. The engineering decoupling point is showing once again that the concept of customer order decoupling point must not be limited to an inventory issue. This decoupling point is quite similar to the information decoupling point in the fact that it uses information from the customer to plan and control the supply chain.

2.6.2. The postponement concept

Postponement is closely linked concept to the CODP. This concept was introduced by Alderson (1950). It is widely recognized as improving supply chain effectiveness (Jones & Riley 1985, Cooper 1993) and it is today a growing trend in manufacturing and distribution (Lampel & Mintzberg, 1996). Logistics postponement consist of delaying value added activities in time and space (Van Hoek, 2001). To this we can add that the final customization is also postponed to the latest as possible (Skipworth & Harrison, 2006). Bowersox and Closs (1996) introduced the three generic types of postponement that are form, time and place postponement. The utilization of the three kind of postponement in the same time is called postponed manufacturing (Van Hoek, 1997).

--- Rijksuniversiteit Groningen

- 41 -

could distinguish several factors that influence the CODP we can find the same factors influencing the postponement: product differentiation, process characteristics, Technological characteristics, market characteristics (Van Hoek, 1997). In fact the ability to postpone will directly influence the location of the CODP. Postponement is generally known as the best way to reach mass customization (Lampel & Mintzberg, 1996. The main benefit of the postponement is mainly that it permits the use of mass customization (Mikkola & Hsuan, 2004).

2.6.3. The modularization concept

Another important concept related to the CODP is the modularization concept. It was introduced by Star (1965) in the literature. The concept aims to marry flexibility with customization by allowing the production of customized product with the same set of standardized component (Ernst & Kamrad, 2000). Baldwin and Clark (1997) distinguish two types of modularization: the modularity in product design and the modularity in production. A modular production process consists of different production modules at different sites that then can be brought together. This allows concurrent production of different sub-process in different sequence (Lee, 1998). The implication for the CODP is that it will shift it downstream because the standardisation of upstream components makes the cost of holding stock less expensive but it also improve the ability of customization. Key issue in this concept is to design efficient products in order to make any combination easily produced.

--- Rijksuniversiteit Groningen

- 42 -

different aspects of the product life cycle: assembly, design, recycling and service (Gu, Hachemian & Sosale, 1999). The main inconvenient of this concept is of course to find the right modular design. This is very costly and difficult.

2.6.4. The leagility concepts

--- Rijksuniversiteit Groningen

- 43 -

Figure 10 . The concept of leagility (Van Hoek, 1999).

The advantage of leagility is a combination of the advantage of both lean manufacturing and agile manufacturing. Upstream the CODP the manufacturing process is characterized by elimination of waste and high productivity while downstream the process the manufacturing process is designed to be able to answer to the customer demand more effectively. This concept also helps to overcome the cost-service trade-off.

2.6.5. The consequence of these theories for the CODP

--- Rijksuniversiteit Groningen

- 44 -

2.7. Conclusion

To summarize the basics of the CODP concept we can determine several characteristics for it:

• CODP position is determined by product, process and market characteristics of an organization.

• CODP can be shifted by changing the product, process and market characteristics. Postponement and modularization concept can be used to change product and process characteristics.

• CODP separates the value chain into a forecast oriented and an order oriented part. Leagility concept can be used to optimize each part.

Aim of this chapter was to give a complete overview of the CODP concept. As we could see the main purpose of the CODP is to decrease costs through a reduction of inventory holding costs. This reduction is acquired with respect to customer requirements. This could be assimilated as a cost-service trade-off situation. In fact as we could see the CODP concept helps to overcome this trade-off. On one hand the positioning of the CODP at the most suitable position will allow to firm to have logistic structure that will fit the best the customer requirement at the lower cost. On the other hand the ability to divide the value chain into two distinct parts will help to optimize the value chain itself and the manufacturing processes by having a strategy that fit the environment (forecast or order oriented).

--- Rijksuniversiteit Groningen

- 45 -

service. Such strategies could be seen as extensions of the CODP concept because the CODP is a compulsory object in order to implement these strategies.

What we can notice through the overview of this subject is that sometimes the existence CODP is not very clear in the literature. Many authors speak about decoupling point that can be assimilated to CODP but that have other names. In order to be clear it is necessary to explain one more time that CODP in our paper correspond to the point where the value chain is divided into two distinct parts : the forecast oriented part and the order oriented part. In our case the CODP correspond also to a physical stock of products unless it is positioned at the very beginning of the manufacturing process.

--- Rijksuniversiteit Groningen

- 46 -

Chapter 3. Discussion about the limitations of the CODP

What we want to discuss now is the limitations that may exists for each of the characteristics we highlighted. Specific processes of the firm, need of multiple CODP or consequences of postponement may lead to difficulties in the application of the CODP concept.

3.1. Problem with determining the position of the CODP

A first limitation that we can see for the CODP concept is that the positioning of the CODP is not always a simple task. As we stated before the position of the CODP depends on product, process and market characteristics. What we can see through the literature is that it is sometimes hard to position a CODP in a precise position. When developing the CODP concept the authors always assume that the value chain can be divided in two distinct parts (Hoekstra & Romme, 1984; Olhager, 2003 & 2006). In reality it is not always the case. The steps to determine the position of the CODP are the following:

• First the organization looks at the market expectations and determines the expected delivery lead time.

• Then according to this lead time they look at how to reach this lead time for the cheapest cost. Mainly inventory holding costs are taken into consideration. • Finally they decide on a position that will reach a good compromise between

the cost of stocks and the expected delivery lead time.

--- Rijksuniversiteit Groningen

- 47 -

3.1.1. Machine setup and batch size

The batch size of a process represents the minimum quantity to produce in one time for a machine in order to reach affordable cost. Batch size usually takes into consideration the setup time of a machine. If the setups are long it is better to have large batches in order to avoid having to set the machine up too often (Schoënsleben, 2001).

At first it is necessary to remind that large batches lead to longer lead time and longer lead time generally has an upstream effect on the CODP position. The machine setup is linked to the same effect. But more than influencing the position of the CODP large batches may prevent from using certain logistic structures like MTO as it is defined in the CODP concept.

What if the downstream production process is using large batch sizes? In this case it will be difficult to produce to order because some information about demand will be missing. The machines have to produce regularly in order not to be idle and to fully use the capacity otherwise costs will increase. The problem is that if demand does not arrive regularly a machine will have to start producing without having received all its orders. If they wait for all orders it will result in a loss of capacity because the machine will be idle .If they produce only the received orders it will result in an increase of costs because of smaller batches and long machine setup. The more the batch is big the more this situation is likely to happen. The decision about what to produce is then based on orders and on forecast. The result is that companies will have end product inventory while they produce on order. The CODP concept is here not well defined because even if the most suitable position leads to a MTO logistic structure it is still not possible to reach it because of large batches. The result will be a hybrid structure between MTO and MTS that is not taken into consideration in the CODP concept.

--- Rijksuniversiteit Groningen

- 48 -

optimum for the firm. It may be impossible to reach the expected delivery lead time if the CODP is moved upstream. And if the CODP is moved downstream one more step of production will be produced to stock and this will result in high inventory cost. This position is in fact the best one but then the concept of CODP must be revisited.

In fact this situation is exactly the situation that was already mentioned by Mason-Jones & Towill (1999). In this case the material decoupling is positioned after the manufacturing step using large batches because from this point all the materials are produced to order. However the information about the customer order is already used one step before to determine the batches.

How to consider the CODP concept in this case? The traditional CODP concept do not include such a situation where the production is half produced to stock and half produced to order. How to combine make to stock and make to order was already an issue mentioned by Van Donk (2001). In his article about food processing industries he highlighted the fact that order received with an amount smaller than the minimum batch size are problematic and affects market strategy. However no solution is given. This is a consequence of the application of the CODP concept that is often regarded as a problem.

What we see here is that CODP cannot be placed anywhere. Hoekstra & Romme (1984) already mentioned this by explaining that CODP position is constrained by the number of planning point that exists along the value chain. We can add that the CODP cannot be placed before a machine using large batches. In case MTO is combined with large batches the CODP concept reaches a limit because the MTO is not anymore a “pure” MTO.

3.1.2. Bottleneck

--- Rijksuniversiteit Groningen

- 49 -

volatile demand and variety of different products. But on the other hand it is best to have the bottleneck downstream the CODP so that the bottleneck only needs to work on products for witch the firm has customer orders, this is the case if the resource of the bottleneck is performing significant activities in the production process of the product (Olhager, 2003). There is today no clear answer about how to position the CODP according to the bottleneck in the CODP concept.

This results again in situation where the CODP position is not very clear. The fact is that if a bottleneck is MTO (CODP is placed upstream) it may happen that the bottleneck is idle because of lack of demand. In theory a bottleneck should not be idle, that means that it should work on MTS situation when there is no demand in order to fully use the capacity. The problem of bottleneck is thus that is creates a hybrid structure that consists of MTO and MTS situations. There is no mention of this structure in the CODP literature.

3.2. Problem with shifting the CODP

There is very little written about how to shift the CODP. “Balancing the diverse factors influencing the location of the decoupling point is missing in the original paper of Hoekstra & Romme as well” (Van Donk, 2001). This part refers to problem in practice encountered with respect to positioning. This will point out some factors that are not present in the literature about CODP concept but that should be taken into consideration.

3.2.1. Adaptations of the systems

--- Rijksuniversiteit Groningen

- 50 -

decides to implement form postponement it is needed to restructure the BOM in order to postpone all the differentiation process to the latest step of the process. This change of the BOM leads to changes in parts acquisition process and manufacturing progress data reporting. With non flexible BOM all these actions have to be done manually because a change of the BOM is not impacted to parts acquisition or reporting. The link between the BOM and parts acquisition and reporting has to be rebuilt. They explain that MRP with configurable BOM are more likely to handle form postponement because all the changes are done instantly. This explains that shifting the CODP has important consequences for information systems and it can not be done in all cases. A limitation of the concept is that it is not possible to shift the CODP with non configurable BOM because of the rigidity of MRP systems.

3.2.2. Form postponement demand excess capacity

--- Rijksuniversiteit Groningen

- 51 -

3.2.3. implication of modularization

Modularization is a good way to manage to shift the CODP downstream. While postponement concerns more the outbound logistics with the way final customization and packaging will be done (Skipworth & Harrison 2005), the modularization is concerned with the inbound logistics and especially with the relationship with the supplier. Several author pointed out the fact that modularization is modifying the buyer-supplier relationship in a supply chain (Doran 2003, Fuorcade & Midler 2005). One key characteristic is the ability of supplier operations to support and drive modular production. Doran (2003) determined a continuum of first tier supplier that distinguishes those suppliers that accommodate change necessary to provide modular solutions and those who don’t. This resulted in a classification of supplier in three categories: fringe suppliers that are poorly positioned to accommodate the changes necessary for modular production; developing suppliers that are developing supplier relationship but that have limited global supply chain presence for one firm; mature suppliers that may control key elements of the supply chain. What is interesting to point out is that it is not only necessary to implement modular production in a firm in order to successfully shift the CODP downstream. There exist factors that are compulsory for success and having a mature supplier is one of them. This may consist of a limitation of the CODP concept by the way that shifting the CODP is constrained by the supplier characteristics.

3.3. Problem with decoupling the flows

3.3.1. Impact of the multiplicity of CODPs