Postponement at Douwe Egberts
Coffee Treatment and Supply
Liquids
- Public Version -
Author: Jelmer Sangers
University of Groningen Faculty Economics & Business
MSc. Business Administration Operations & Supply Chains
April 2009
Hofstraat 12 9712 JB Groningen jelmersangers@hotmail.com
Student number: 1654942
Organization: Douwe Egberts Coffee Treatment and Supply Company Supervisor: R. Oud
I. PREFACE
This master thesis is the result of my research at Douwe Egberts Coffee Treatment and Supply. Furthermore, it marks the end of my study Master of Science Business Administration, with the specialization in Operations and Supply Chains, at the University of Groningen.
I would like to thank my company supervisor, Ric Oud, for all the support, feedback and giving me the opportunity to do the research for my master thesis at Douwe Egberts Coffee Treatment and Supply. As well as all the employees of Douwe Egberts who helped me during this research, and especially the employees of the planning and logistics department, for their support. Furthermore, I would like to thank my supervisors of the University of Groningen, Gera Welker and Arnaud Pool, for the support and feedback. In addition, some special thanks to my parents for allowing me to stay at home again during this research project, which saved me a lot of travel time, their great care and healthy meals.
This research was a great experience which gave me the opportunity to put the theoretical knowledge of my study into practice. I hope you will enjoy reading my master thesis, as much as I enjoyed the research and writing of my master thesis.
Jelmer Sangers
II. EXECUTIVE SUMMARY
The objective of this research was to see if it would be viable for DE CT&S to postpone the customization of the Cafitesse products till after the freezing process at Diepvriesveem. Therefore, the following research question was formulated:
“What are the benefits and costs for DE CT&S in case the customization of the Cafitesse products is postponed till after the freezing process?”
Postponement is associated with lower inventories, better service, and lower operational costs. Therefore, the focus has been on these points. However, first the most suitable postponement strategy had to be determined. Based on the characteristics of the Cafitesse products, a packaging/labeling postponement strategy would be best suitable. Whereby, the Cafitesse products would be kept in a generic state after the freezing process and are customized with packaging/labeling when an order is received. In order to achieve this, several packaging concepts are discussed during this research; sleeve, wrap foil, sticker, bundle, and duo-pack. These package concepts represent the package material which contains the customer specific information.
With a packaging/labeling postponement concept DE CT&S would be able to reduce the average unrestricted use inventory of the customized BiBs with 24% in case of the 1.25 litre BiBs, and with 38% in case of the 2 litre BiBs, if the BiBs are kept in stock in a generic form. The inventory level of broadsheet would also be reduced if kept in a generic form. However, these savings are offset by the cost of having extra package material in stock for making the BiBs customer specific. Therefore, the savings made in the inventory of package material are minimal and could also lead to extra costs dependant on the packaging concept.
customers. Hence, the delivery lead time to these customers will increase. Furthermore, the Cafitesse will be delivered from stock with generic BiBs which already have passed the quality inspection. Hence, with a packaging/labeling postponement strategy the failure in quality inspection of a Cafitesse product will no longer influence the delivery lead time to the customer.
The operational costs are not expected to decrease with a packaging/labeling postponement strategy for several reasons. The savings in set-up time of the current packaging process are expected to be minimal. Additional employees are required to operate packaging processes at two locations, DE CT&S and Diepvriesveem, which is expected to be 1.5 FTE. Furthermore, investments are needed in new machinery, which are dependent on the packaging concept. These machines require space at Diepvriesveem, which will change the lay-out at Diepvriesveem resulting in extra costs. As well as additional warehouse space is needed at Diepvriesveem in order to store the package material needed for the packaging processes at Diepvriesveem, with a packaging/labeling postponement strategy, resulting in extra storage infrastructure costs.
III. GLOSSARY
Agile production : Building up of potentials or scope in the right time and in the right amount.
Average unrestricted
use inventory : Inventory which passed quality control and is ready to be used. BiB : Bag in Box, ‘flexizak’ with liquid coffee packed in a broadsheet. Broadsheet : A publication consisting usually of a single sheet bearing matter printed
on both sides of the sheet; Dutch: plano.
Cafitesse products : Liquid coffee extract used in coffee machines. CODP : Customer Order Decoupling Point.
DECS : Douwe Egberts Coffee Systems.
DE CT&S : Douwe Egberts Coffee Treatment and Supply.
Delivery lead time : The total time required to receive, fill, and deliver an order; the time from receipt of a customer order to the delivery of the product or the fulfilment of the service.
EC : European Collaborative.
Fill rate : Number of products delivered on desired delivery date divided by number of products ordered.
Flexibel bag : ‘Flexizak’, package material for liquid coffee extract.
FTE : Full Time Employment.
IQ blend : Improved Quality.
Lead time : Span of time required to perform a process (or a series of operations). Lean production : The minimization of all required resources (including time) for the
various activities of the company. It involves identifying waste and eliminating them.
MTO : Make-to-order.
MTS : Make-to-stock.
NEC : Non-European Collaborative. NENC : Non-European non Collaborative.
OpCo’s : Operating Companies.
SKU : Stock Keeping Unit.
SO : Sales order.
TQL blend : Top Quality Liquid.
ZFIN : Finished product.
IV. TABLE OF CONTENT
I. PREFACE 2
II. EXECUTIVE SUMMARY 3
III. GLOSSARY 5
IV. TABLE OF CONTENT 6
1. COMPANY PROFILE 9
1.1 History of Douwe Egberts and Sara Lee Corporation 9
1.2 Douwe Egberts Coffee Treatment and Supply 9
1.3 Products of Douwe Egberts Coffee Treatment and Supply 10
2. RESEARCH DESIGN 11 2.1 Initial Move 11 2.2 Problem Statement 11 2.2.1 Research Objective 11 2.2.2 Research Question 12 2.3 Theoretical Framework 12 2.3.1 Postponement Strategies 12
2.3.2 Effects Postponement Strategy on Inventory 14
2.3.3 Effects Postponement Strategy on Service Level 15
2.3.4 Effects Postponement Strategy on Operational Costs 15
2.4 Conceptual Model 16
2.5 Sub-questions 16
2.6 Research Methodology 17
3. PRODUCTION PROCESS, PRODUCT AND MARKET 18
3.1 Current production process 18
3.1.1 Roasting and Extraction 18
3.1.2 Packaging 19
3.1.3 Freezing and Quality Control 19
3.2 Product Characteristics 19
3.2.1 Commonality of Modules 20
3.2.2 Product Variety 21
3.3 Market Characteristics 23
3.3.1 Markets 23
3.3.2 Product Life Cycle 24
3.3.3 Sales Fluctuation 24
3.3.4 Lead Time 25
3.3.5 Price 26
3.4 Conclusion 26
4. POSTPONEMENT AT DOUWE EGBERTS COFFEE TREATEMENT AND
SUPPLY ` 27
4.1 Postponement Strategy for the Cafitesse Products 27
4.2 Conclusion 29
5. REDESIGN OF PRODUCTION PROCESS AND PRODUCT 30
5.1 Customer Order Decoupling Point 30
5.1.1 Customer Order Decoupling Point before Postponement 30
5.1.2 Customer Order Decoupling Point after Postponement 32
5.2 Packaging Concepts for the Postponement Strategy 33
5.3 Conclusion 34
6. INVENTORY 36
6.1 Average Inventory and Safety Stock Calculation 36
6.1.1 Overview Current and Expected Safety Stock and Average
Inventory 39
6.1.2 Inventory Turnover Ratio 41
6.2 Inventory Levels Packaging Materials 42
6.3 Conclusion 44
7. SERVICE 45
7.1 Delivery Lead Time 45
7.2 Fill Rate 47
7.3 Conclusion 48
8. INVESTMENTS, COSTS, AND SAVINGS 49
8.1 Investments in Packaging Process 49
8.1.1 Complexity of Final Manufacturing Process and Technological
Content 52
8.2 Savings in Set-up Time 53
8.4 Savings Package Material 57
8.5 Costs Associated with Leakages 58
8.6 Summary 59
9. CONCLUSION AND RECOMMENDATION 61
9.1 Conclusion 61
9.2 Recommandation 63
9.3 Reflection 64
V. REFERENCES 65
VI. APPENDICES 67
1. Overview OpCo’s Liquids 68
2. Demand per Week P01FY08 – P03FY09 69
3. Lead Times 71
4. European Guideline 2000/13/EC 72
5. Figures used for Calculating Generic BiBs Inventory Levels 73
6. Inventory Turnover Ratio Calculation 74
7. Inventory Levels Broadsheets 1.25 Litre 75
8. Inventory Levels Broadsheets 2 Litre 76
9. Figures used for Calculating Broadsheets Inventory Levels 77
10. Overview Planning VL08 Week 40 78
11. Expected Inventory Levels 1.25 Litre Packaging Materials 79
12. Expected Inventory Levels 2 Litre Packaging Materials 80
13. Savings Holding Cost with Generic BiB Inventory 1.25 Litre 81
14. Savings Holding Cost with Generic BiB Inventory 2 Litre 82
15. Expected Costs Package Material Packaging Concepts 83
16. Total Cost of Package Material Waste when Product Fails Quality Inspection 85
1. COMPANY PROFILE
This chapter introduces Douwe Egberts and Sara Lee Corporation. The organizations and their history and Douwe Egberts Coffee Treatment and Supply (DE CT&S) are described in this chapter. As well as the products produced by DE CT&S.
1.1 History of Douwe Egberts and Sara Lee Corporation
Douwe Egberts started in 1753 when Egbert Douwes opened a store in Joure, “De Witte Os”, in which he sold goods which belonged to the delights of daily life. Coffee, tea, and tobacco were the most important products. In 1780 his son, Douwe Egberts, started to work in the store. Since that time the company became known as Douwe Egberts. In 1919 Douwe Egberts expanded from a regional company to a national company by opening a store in Utrecht. In 1978 the family-owned company ended when Douwe Egberts started to cooperate with the American company Consolidated Foods, later known as Sara Lee. After this cooperation and some acquisitions Douwe Egberts and Sara Lee joined together in 1984 under the name Sara Lee/DE. In 1989 Sara Lee/DE acquired Van Nelle, including Lassie. Both brands were placed under Douwe Egberts Netherlands.
In 2005 Sara Lee Corporation started to transform their organization, focussing on strengthening their focus on food, beverage, and household and body care markets. Therefore, the identity of the whole organization is integrated in one logo, the Sara Lee logo. As result of this the name Sara Lee/DE is changed in Sara Lee International.
1.2 Douwe Egberts Coffee Treatment and Supply
Logistic Department follows from the mission of DE CT&S: “Providing planning schedules for supplying, production and delivery purposes to achieve an accurate and timely delivery of coffee products to customers.”
Figure 1.1: Organization chart DE CT&S.
1.3 Products of Douwe Egberts Coffee Treatment and Supply
The instant coffee produced by DE CT&S is used to create a cup of coffee by dissolving the instant coffee in hot water. The liquid coffee is used in coffee machines, delivered by Douwe Egberts Coffee Systems (DECS). The products that are produced by DE CT&S are divided in four product groups:
• Spray (instant), • Agglo (instant), • Freeze-dry (instant) • Liquid.
These products are packed and send to the sales organizations of Sara Lee all over the world. Some of the well known product brands that are produced by DE CT&S are: Douwe Egberts, Moccona, Merrild, Maison du café, Marcilla, Harris, and Café do Ponto.
2. RESEARCH DESIGN
This chapter describes the initial move, the problem statement, and the theoretical framework of this research. Furthermore, the sub-questions, the conceptual model, and the research methodology of this research are described.
2.1 Initial Move
DE CT&S produces liquid coffee extract for coffee machines, Cafitesse, for customers all over the world. Currently, DE CT&S is producing ten different liquid coffee blends and one liquid tea blend in 1.25 and 2 liter units. In addition these units have different packaging depending on the country towards they are shipped. Currently, there are around 100 different kinds of finished Cafitesse articles. These products are produced on order and to stock depending on the rate of sale, fast movers and slow movers. The lead time is four to five weeks depending on the type of extraction method and can increase if the product fails quality inspection. Currently, the Cafitesse products are customer specific packed and labeled at DE CT&S after which they are transported to Diepvriesveem where the Cafitesse products are frozen and restacked. At this time DE CT&S is not facing major problems with the production of Cafitesse products. However, if DE CT&S is able to apply a postponement strategy, would the benefits of a postponement strategy (lower inventories, better service, and lower costs), outweigh the cost of implementing a postponement strategy.
2.2 Problem Statement
Based on the initial move of this research the research objective and research question are formulated in this paragraph.
2.2.1 Research Objective
2.2.2 Research Question
The research question of this research is: “What are the benefits and costs for DE CT&S in case the customization of the Cafitesse products is postponed till after the freezing process?”
2.3 Theoretical Framework
This paragraph will describe the theoretical background of postponement strategies, the benefits and costs associated with postponement strategies.
2.3.1 Postponement Strategies
Postponement, or late customization, is a product design approach that shifts product differentiation closer to the customer by postponing identity changes, such as assembly or packaging to the last possible facility location (Schönsleben, 2007) and is seen as an important supply chain strategy which is more and more used over the last years. Van Hoek (2001) describes postponement as: “delaying activities in the supply chain until customer orders are received with the intention of customizing products, as opposed to performing those activities in anticipation of future orders”. In the literature there are many reasons mentioned for the increasing importance of postponement strategies, which is seen as one of the most beneficial strategies for reducing the costs and risks of product variety and improving the performance of the supply chains (Davila and Wouters, 2007).
Figure 2.1: Postponement strategies and positioning the customer-order decoupling point (source: Yang et al., 2004a)
Van Hoek (2001) has made an overview of the operating characteristics that are relevant to assess the viability of a postponement strategy. The operating characteristics are divided in: product, market, process, and technology characteristics. Furthermore, Van Hoek (2001) describes how these characteristics relate to postponed manufacturing (table 2.1). In the literature several benefits of postponement can be found. These are lower inventory levels, increased service levels, and lower operational costs (Davila and Wouters, 2007). The benefits in terms of lower inventory, faster response times to customers, increased availability levels and increased flexibility, are difficult to quantify. Since a future situation has to be described and calculated which does not yet exist. It is here that inventory theories and models can make a contribution (Lee, 1996).
Product Characteristics (Chapter 3) Effects of Postponement
High commonality of modules (paragraph 3.2.1) Lowered inventory levels and reduced risk of obsolete inventories
Product variety; specific formulation of products (paragraph 3.2.2)
Improved customization
Product variety; specific peripherals/packaging (paragraph 3.2.3)
Improved customization
Product cube and/or weight increases through customization/final manufacturing (paragraph 3.2.5)
Reduced transportation and inventory carrying costs
Market Characteristics Effects of Postponement
Short product life cycle (paragraph 3.3.2) Less risk of obsolete inventories
High sales fluctuation (paragraph 3.3.3) Reduced inventory levels and less risk of obsolete inventories
Short and reliable lead times required (paragraph 3.3.4) Improved delivery service Price competition (paragraph 3.3.5) Lowered cost levels Varied and (physically) fragmented markets (paragraph
3.3.1)
Better targeting, segmentation, and positioning of product and sales
Process Characteristics Effects of Postponement
Possible to decouple primary and secondary production system (paragraph 5.1)
(a technical pre-condition and needed for manufacturing within lead time)
Limited complexity of final manufacturing process (paragraph 8.1.1)
Short set-up and changeover times, short processing times
Sourcing from multiple locations Direct bulk shipments of modules
Technology Characteristics Effects of Postponement
Limited complexity of final manufacturing operation (paragraph 8.1.1)
Limited loss of economies of scale through postponement and short processing times
Limited complexity of technological content in final manufacturing (paragraph 8.1.1)
Short set-up and changeover times, short processing times
Modularity (paragraph 3.2.1) Rapid final manufacturing at low processing costs, increased possibility to adjust products to markets
Table 2.1: Overview of the characteristics and the effects of postponement (source: Van Hoek, 2001)
2.3.2 Effects Postponement Strategy on Inventory
increase of inventory turns lowers the risk of inventory becoming obsolete. The effects of the postponement strategy on inventory will be measured based on expected inventory levels and inventory turnover ratio after the implementation of a postponement strategy.
2.3.3 Effects Postponement Strategy on Service Level
Another benefit associated with postponement strategies is improved service level. Service level is mostly measured with the fill rate, number of products delivered on desired delivery date divided by number of products ordered (Schönsleben, 2007). If the postponed final manufacturing is still able to deliver within the required delivery lead time, the improved responsiveness may save on backorder and increase the fill rate, thereby improving the delivery reliability and service (Van Hoek, 2001). However, when additional cycle time needed for the final manufacturing is longer than the required delivery lead time, the service level will decrease, in which case postponement is not required. Delivery lead time is the total time to receive, fill, and deliver an order; the time from receipt of a customer order to the delivery of the product or the fulfillment of the service (Schönsleben). The effects of the postponement strategy on the service level will be measured based on the expected delivery lead time and fill rate after the implementation of postponed manufacturing.
2.3.4 Effects Postponement Strategy on Operational Costs
2.4 Conceptual Model
The conceptual model is designed based on the problem statement and the theoretical framework. The benefits associated with postponement strategies are: lower inventories, increased service level, and lower operational costs. However, first the viability of a postponement strategy will be assessed with the market and product characteristics. After assessing the viability, a postponement strategy will be chosen based on the product and market characteristics of the Cafitesse products. The product design has to be re-designed to implement a postponement strategy. Furthermore, the production process will change due to the implementation of a postponement strategy. However, this is dependent on processes and technology characteristics and if they enable the re-design of the production process. All these issues will be dealt with when answering the sub-questions. Figure 2.2 gives an overview of the conceptual model.
Figure 2.2: Conceptual model
2.5 Sub-questions
Based on the conceptual model the following sub-questions are formulated in order to answer the research question and thereby reach the objective of this research:
1. What are the product characteristics, and how do they influence the effects of a postponement strategy?
Choice of postponement strategy (Chapter 4)
Expected performances for the Cafitesse product with a form
postponement strategy
Inventories (Chapter 6): • Stock levels
• Inventory turnover ratio Service level (Chapter 7): • Lead time
• Fill rate
Operational costs (Chapter 8): • Investments • Costs • Savings Cafitesse characteristics (Chapter 3) Product characteristics: • Commonality of modules • Product variety • Value density • Product weight Market characteristics: • Product life cycle • Sales fluctuation • Lead times • Price competition • Markets
Redesign of production process and product (Chapter 5): • Process
characteristics • Technology
2. What are the market characteristics, and how do they influence the effects of a postponement strategy?
3. Which postponement strategy fits best with the situation of the Cafitesse product?
4. What kind of redesign of the product and production process is necessary in order to implement the proposed postponement strategy and how does this affect the CODP? 5. How does the proposed postponement strategy affect the inventories?
6. How does the proposed postponement strategy affect the service level? 7. How does the proposed postponement strategy affect the operational costs?
8. What are the savings, investments, and costs associated with the redesign of the product and production process?
2.6 Research Methodology
In order to describe the characteristics of the Cafitesse product and market (sub-questions one and two) discussions were held with the employees of logistic supply. Based on the description of the characteristics, their influence on the effects of a postponement strategy were described based on literature review.
To determine the most suitable postponement strategy for the Cafitesse product (sub-question three) the models of Pagh and Cooper (1998), and Yang et al. (2004b) where used.
The necessary redesign of the Cafitesse product and production process (sub-question four) is determined based on discussion with the package technologist and packaging engineer.
In order to determine the effects of a postponement strategy on the inventory levels, service level and operational costs (sub-questions five, six and seven) data provided by DE CT&S is used as well as theoretical models found in literature.
3. PRODUCTION PROCESS, PRODUCT AND MARKET
This chapter describes the current production process. Furthermore, an analysis is made of the product characteristics and the market characteristics, mentioned in table 2.1, of the Cafitesse products. Thereby, it will answer the first and second sub-question of this research:
• What are the product characteristics, and how do they influence the effects of postponement strategy?
• What are the market characteristics, and how do they influence the effects of postponement strategy?
3.1 Current production process
The production process starts with the roasting of the coffee beans, after which the extraction process starts. The extraction process results in ten liquid coffee blends for the Cafitesse products. These blends are packed and frozen. Figure 3.1 gives a schematic overview of the production process. The different processes will be described in the next sub-paragraphs, see Van Wieren (2006) and Pool (2007) for a more detailed description of the production process.
Figure 3.1: Production process
3.1.1 Roasting and Extraction
The production process starts with the blending and roasting of green coffee beans. These roasted and blended coffee beans are the input for the extraction process. After the extraction process ten different liquid coffee blends are available for the Cafitesse products. The roasting and extraction process will not be described in detail since the research focuses on the packaging process of the Cafitesse products.
Location OpCo Diepvriesveem
DE CT&S
3.1.2 Packaging
After the extraction, the ten different coffee blend extracts are packed. The extract is packed in volumes of 1.25 and 2 litres in folded and sealed foil which is fitted with a dosing system. These are placed on a broadsheet which is folded and glued together, which results in a so-called Bag-in-Box (BiB). Finally, the BiB is put in a flow wrap. The flow wrap prevents, in case of a leakage, that other products become stained. The BiB is placed in trays in quantities of two and tied with straps, 1.25 litre products can also be placed in trays in quantities of four. These packages are placed on a plastic pallet with space between the packages in order to stimulate the freezing process. These articles get a so called ZSEM article number, which is the article number for Cafitesse products ready to be frozen. All these products are shipped around every three hours to Diepvriesveem where they are frozen and stored. During the packaging process samples of each batch are taken for quality control.
3.1.3 Freezing and Quality Control
At Diepvriesveem the product will be frozen. After the freezing process, samples of each batch of Cafitesse products are taken for quality control and the Cafitesse products with leakages are removed. At Diepvriesveem the Cafitesse products are restacked on new pallets and stickers applied according to customer requirements. Stickers are only applied on Cafitesse products to customize small orders. When the products are restacked the Cafitesse products get their final article number, ZFIN. The ZSEM and ZFIN products are mostly one-on-one related. However, the stickers and different pallet stacking result in a small increase of ZFIN products with regard to the ZSEM products. When the Cafitesse passes the quality control it can be shipped to the customer after an order is received. Cafitesse products that do not pass the quality control are send back to DE CT&S and are partly reused.
3.2 Product Characteristics
3.2.1 Commonality of Modules
A modular product concept is based on standardizing the components and operations as well as building product families (Schönsleben, 2007). Product modularity enables fast assembly and cost efficiency required for postponement strategies. When the modules which are used in the final product have a high commonality, the level of inventory can be lowered and the risk of obsolete inventories reduced. Due to increasing inventory turnover and aggregate sales forecasts (Levi et al., 2008).
The Cafitesse products have several common materials. All the Cafitesse products consist of a blend and package materials. There are a total of around 100 final (ZFIN) Cafitesse products at the moment. Figure 3.2 gives an overview of the bill of material of a Cafitesse product. However, the materials consist of many variants which lead to the many different final Cafitesse products. Table 3.1 gives an overview of the number of variants per material used in the final Cafitesse products. When a postponement strategy is applied to the Cafitesse products, whereby the Cafitesse product is customized after the freezing process, the packaging materials used before the freezing process will be common for each blend. This will result in generic 1.25 and/or 2 litre modules per blend. These can be customized after a customer order is received. This will give DE CT&S the opportunity to reduce inventory levels and the risk of obsolete inventories.
Figure 3.2: Bill of Material Cafitesse.
Cafitesse
Blend Package material
Wrap foil Straps
Foil
Broadsheet
Dosing system
Glue Stickers Flexible bag Trays
1.25 & 2 L
Pallet Pallet label Pallet with Cafitesse
Material # of variants
Glue 1
Wrap foil 2 (1,25 & 2 litre)
Straps 1
Pallet 2 (euro & slipsheet)
Foil 1
Dosing system 2 (IQ & TQL)
Tray 27
Sticker (excl. pallet sticker) 15
Broadsheet 48
Blend 10
Table 3.1: Overview of number of material variants (Based on Cafitesse products sold between P01FY08 and P02FY09)
3.2.2 Product Variety
Blend Volume # ZFIN products 605 1.25 & 2 litre 44 612 1.25 & 2 litre 12 613 1.25 & 2 litre 3 618 1.25 & 2 litre 6 631 1.25 & 2 litre 10 646 1.25 & 2 litre 5 651 1.25 & 2 litre 2 672 2 litre 4 685 1.25 & 2 litre 3 691 2 litre 8
Table 3.2: Volumes and ZFIN products per blend
Figure 3.3: Schematic overview of increase of variety during the packaging process
Figure 3.4: Number of customers per ZFIN product
Postponed manufacturing delays the customization of a product. Therefore, a high product variety in terms of specific formulation and peripherals allows for improved customization when adapting a postponement strategy. Hence, instead of having Cafitesse products which
68% 17% 6% 2% 7% 1 Customer 2 Customers 3 Customers 4 Customers 5 or more Customers Packaging Freezing
Extract ZSEM ZFIN
97 91
can be send to multiple customers, DE CT&S could choose to design the packaging material, which makes the Cafitesse product customer specific, in such a way that each final Cafitesse product is specific for one customer.
3.2.3 Value Density of the Products
The value density is the division of unit costs by the size or weight of the object, which is the product value per kilogram or cubic meter (Schönsleben, 2007). If the value density of the product is high, postponed manufacturing reduces the pipeline expenses and inventory carrying costs. Since, postponed manufacturing reduces the level of inventory. The value density of a Cafitesse product is between €X and €X per litre. A postponement strategy enables DE CT&S to reduce the inventory of the Cafitesse products. Hence, a reduction can be made in pipeline expenses and inventory carrying cost.
3.2.4 Product Cube and/or Weight Increase through Customization
When the product cube and/or weight increases through final manufacturing transportation volume and storage space needed increases, postponed manufacturing can reduce the costs of transportation and inventory costs. However, it is not expected that there will be an increase in the product cube and/or weight of the Cafitesse products. Since, the products are designed to fit in the Cafitesse machines all around the world. Changing the product cube and/or weight would require replacing all these machines which is not an option. Furthermore, the final manufacturing will consist of packaging and/or labeling of the Cafitesse product which will not have a huge impact on the product cube or weight. Taking this in consideration it is not expected that there will be a reduction in transportation and inventory carrying costs.
3.3 Market Characteristics
This paragraph describes and analyses the market characteristics of the Cafitesse product and how these characteristics relate to the effects of a postponement strategy. The characteristics that will be described are; markets, product life cycle, sales fluctuation, lead time, and price.
3.3.1 Markets
organizations within Sara Lee, the so called OpCo’s. These organizations deliver the Cafitesse products to the out-of-home markets all around the world. The customers can be divided in four different types of customers:
• European collaborative (EC) • Non European collaborative (NEC) • Non European non collaborative (NENC) • Sales order (SO)
For this research the different types of customers are not described in detail, Van Wieren (2006) gives a more detailed description of the customers. Appendix 1 gives an overview of the OpCo’s and the type of customers they are. Hence, the targeted markets for the Cafitesse products are varied and physically fragmented. In these situation postponement strategies allow for better targeting, segmentation, and positioning of the product and sales.
3.3.2 Product Life Cycle
Schönsleben (2007) defines product life cycle as: the market stages a new product goes through from the beginning to end, i.e. introduction, growth, maturity, saturation and decline. Short product life cycles increase the risk of inventories becoming obsolete. Postponement lowers that risk, especially when the components have a high commonality and can be used in other products (Van Hoek, 2001). The Cafitesse product can be described as a functional product. Functional products are characterized by long product life cycle, contrary to innovative products which are characterized with short product life cycle (Levi et al., 2008). Hence, the Cafitesse product does not have a short product life cycle. Furthermore, it is noticed that the blends produced by DE CT&S are stable. However, some IQ blends are replaced by TQL blends, e.g. blend 612 will be replaced by blend 613. However, this is done in such a way that it does not lead to obsolete inventories. Based on the fact that the Cafitesse products have a long product life cycle, it is not expected that a postponement strategy will result in less risk of obsolete inventories for DE CT&S.
3.3.3 Sales Fluctuation
quite stable, since no special actions are being held in these markets. Only for Cafitesse products that are used in the holiday sector seasonal influences are visible. An overview of the sales per week from P01FY08 up to P03FY09 included can be found in appendix 2. As already mentioned earlier the Cafitesse products can be described as a functional product. One of the characteristics of a functional product is that demand is stable and predictable (Levi et al., 2008). When there are high sales fluctuations postponement strategies will reduce the inventory levels and lower risk of obsolete inventory. Since, the aggregate forecasts of the modules are more accurate which results in less uncertainty in demand leading to reduction in safety stock (Levi et al., 2008). This increases inventory turns reducing the risk of obsolescence. However, this is not the case for the Cafitesse products where the demand is quite stable and predictable. Therefore, based on this characteristic reduced inventory levels and lower risk of obsolete inventory is not expected.
3.3.4 Lead Time
3.3.5 Price
DE CT&S produces the Cafitesse products for the OpCo’s within Sara Lee. Hence, DE CT&S is seen as a manufacturing unit. The goal is to supply the OpCo’s Cafitesse products against a cost price which is as low as possible. Although, DE CT&S is not really competing on price, a low cost price is important. The postponement strategy should result in a lower cost price per unit. It is possible that the postponement strategy results in a trade-off between different cost units. However, the overall cost price per unit should be lower.
3.4 Conclusion
With the findings described in this chapter the first and second sub-questions of this research can be answered:
• What are the product characteristics, and how do they influence the effects of postponement strategy? Based on Van Hoek (2001) the product characteristics that influence the effects of postponement have been determined. The high commonality of modules in the Cafitesse products enables to lower the inventory levels and reduce the risk of obsolete inventories with a postponement strategy. The high product variety in terms of specific formulation of products and peripherals/packaging enable to improve the
customization of the products with a postponement strategy. However, the value density of the Cafitesse products and the lack of product cube and/or weight increases through customization/final manufacturing, reduce the effects a postponement strategy can have on reducing pipeline expenses, transportation, and inventory carrying costs.
4. POSTPONEMENT AT DOUWE EGBERTS COFFEE TREATEMENT AND SUPPLY
This chapter discusses which postponement strategy is most suitable for the Cafitesse products. Thereby answering the third sub-question of this research:
• Which postponement strategy fits best with the situation of the Cafitesse product?
4.1 Postponement Strategy for the Cafitesse Products
In order to identify a suitable postponement strategy for DE CT&S the model described by Yang et al. (2004b) is used (figure 4.1). Yang et al. (2004b) identify two levels of uncertainty: low and high level of uncertainty. Low uncertainty is uncertainty about the demands at a single product level (while aggregate demands can be more accurately forecasted), or uncertainty as to where and to whom finished products should be distributed (when there is enough information to initiate all production activities). High uncertainty is uncertainty as to what, when and how many to produce. Yang et al. (2004a) relate the degree of uncertainty with the degree of modularity in the product. Modularity is building a product from smaller subsystems that can be designed independently yet function together as a whole (Yang et al. 2004a). In the case of DE CT&S and the Cafitesse products there is a low uncertainty, regular updates of the forecasts by the OpCo’s and stable demand for the out-of-home market, and as discussed in paragraph 3.1.1 there is a high degree of product modularity. Hence, production postponement, or postponed manufacturing is best suitable for the situation at DE CT&S and the Cafitesse products.
Figure 4.1: Model for the management of uncertainty through postponement (source: Yang et al. 2004a)
postponed manufacturing strategies is best suitable for the Cafitesse? Based on simple “yes” or “no” questions related to the characteristics of the product Pagh and Cooper (1998) identify four postponement strategies, bundled manufacturing, unicentric, deferred assembly, and deferred packing (figure 4.2). By answering the simple “yes” or “no” questions of Pagh and Cooper (1998) the best suitable postponed manufacturing strategy for DE CT&S can be identified. The first question to be answered is: “Is the formulation common to all markets?”. In the case of the Cafitesse product this question is answered with a “yes”. The second question to be answered is: “Are the peripherals (label, packaging, instruction manual) common to all markets. This question is answered with a “no” in the case of the Cafitesse product. Hence, using the matrix in figure 4.2 it can be concluded that a deferred packaging strategy is best suitable in the case of the Cafitesse products.
Figure 4.2: The basic concept of Cooper’s key work (source: Pagh and Cooper, 1998)
The deferred packaging strategy as described by Pagh and Cooper (1998) relates strongly to the packaging/labeling postponement strategy of Yang et. al. (2004b), which is seen as form postponement strategy as mentioned by Van Hoek (1997). Table 4.1 gives an overview on which postponement strategies mentioned in different researches relate to each other.
Table 4.1: Overview relation of postponement strategies
Van Hoek (1997) x x Form postponement Time
postponement
Place postponement Pagh and Cooper
(1998)
x x Deferred assembly Deferred packaging Unicentric x
4.2 Conclusion
With the findings of this chapter the third sub-question of this research can be answered: • Which postponement strategy fits best with the situation of the Cafitesse product? Based
5. REDESIGN OF PRODUCTION PROCESS AND PRODUCT
This chapter describes the necessary redesign of the production process and product. As well as the effect of packaging/labeling postponement strategy on the CODP. Thereby answering the fourth sub-question of this research
• What kind of redesign of the product and production process is necessary in order to implement the proposed postponement strategy and how does this affect the CODP?
5.1 Customer Order Decoupling Point
The CODP decouples the production process in a primary and secondary production system. The primary production system is fully forecast driven, which allows it to focus on large economic runs. The secondary production system produces is fully order driven, focusing on responding to customer wishes (Van Hoek et. al. 1998). Hence, the primary production system can be lean production and the secondary production system can be agile. Schönsleben (2007) defines lean production as the minimization of all required resources (including time) for the various activities of the company. It involves identifying waste and eliminating them. Agile manufacturing refers to the building up of potentials or scope in the right time and in the right amount (Schönsleben 2007). With a postponement strategy there is a combination of order driven and forecast driven processes (Van Hoek, 2001). Therefore, the decoupling of the production process in a primary and secondary production system is seen as a pre-condition for implementing a postponement strategy, which is possible for the production process of the Cafitesse products. Paragraph 5.1.1 will describe where the current CODP decouples the production process in the primary and secondary production system. Paragraph 5.1.2 describes where the production process of the Cafitesse products would be decoupled in a primary and secondary production system with a packaging/labeling postponement strategy.
5.1.1 Customer Order Decoupling Point before Postponement
in the demand during the lead time (Schönsleben, 2007). The ‘slow movers’ products are made-to-order and are destined for NENC and SO customers and do not have safety stock (figure 5.1). The destination of the ‘slow movers’ are fixed and cannot be assigned to another customer. Currently, there are around 30 ‘fast movers’ and around 70 ‘slow movers’ Cafitesse products (figure 5.2). Based on the sales volume between P03FY08 and P02FY09 we can conclude that 20% of the sales consist of ‘slow movers’ and 80% of ‘fast movers’ (table 5.1).
Figure 5.1: CODP in production process before postponement
29%
71%
Slow movers
Fast movers
Figure 5.2: Percentage ‘fast movers’ and ‘slow movers’
Table 5.1: Sales volume of current slow movers and fast movers between P03FY08 and P02FY09 (see appendix 17)
Location OpCo Diepvriesveem
DE CT&S
Roasting Extraction Packaging
Freezing + Quality ZFIN: Finished product Coffee beans Package materials CODP: Articles made to order
CODP: Articles made to stock Forecast driven for MTS
5.1.2 Customer Order Decoupling Point after Postponement
The CODP with a postponement strategy is typically located at the non-customer specific product stage such that (Harrison and Skipworth, 2008):
• No previously added value should be removed during the postponed process. There must be no removal of components or rework.
• Postponed value added processing time must be short compared to the total value adding process time required to manufacture the product.
• The number of generic product variants should be kept to a minimum.
When a postponement strategy is applied whereby the packaging/labeling is postponed, the CODP for all the articles will be after the freezing and quality inspection (figure 5.3). Hence, the products that fail the quality inspection will not enter the postponed process (packaging/labeling), which corresponds with the first point mentioned by Harrison and Skipworth (2008). The processing time of the postponed process is dependent on the capacity and speed of the final process. However, it is expected that the postponed processing time will take a maximum of one week (paragraph 7.1), which is shorter compared with the total value adding time of three weeks. This corresponds with second point of Harrison and Skipworth (2008). Placing the CODP after the freezing process allows the roasting, extraction, packaging, and freezing and quality to focus on lean production. See Pool (2007) for lean production planning in the liquids. Besides the benefit of more processes that can be organized lean, there are several other benefits for DE CT&S when postponing the labeling/packaging of the Cafitesse products. These benefits will be discussed in the next chapters.
Figure 5.3: CODP in production process after postponement
Location OpCo Diepvriesveem
DE CT&S
Roasting Extraction Packaging
Freezing + Quality
Final Packaging
Forecast driven Order driven
5.2 Packaging Concepts for the Postponement Strategy
When a packaging/labeling postponement strategy is applied to the Cafitesse products, new modular packaging concepts will have to be designed. The modular design enables rapid final manufacturing at low processing costs and increases the possibility to adjust the products to the market. The following packaging solutions for making the product customer specific are considered in this research, based on discussion with the package technologist:
• Sleeve, • Wrap foil, • Sticker, • Bundle, • Duo-pack.
Table 5.1 gives an overview of the packaging concepts mentioned in this research, sleeve, wrap foil, sticker, bundle, and duo-pack. These represent the final packaging material that will be applied to customize the generic products, and will contain customer specific information. Appendix 4 gives an overview of the information the final packaging material legally must contain according to European guidelines. Furthermore, additional country specific legal information may be required. Table 5.2 also gives an indication on how the packaging concepts score on charisma (attractiveness to the customer), technical feasibility (if it is technically possible to package the product with the concept) , and sustainability (the impact on the environment), based on discussion with the package technologist.
Concepts Description Charisma Technical feasibility
Sustain- ability
Sleeve Generic BiB is sleeved, whereby the sleeve contains all the
required information.
++ +/- -
Wrap foil Generic BiB is placed in a wrap foil, whereby the wrap foil
contains all the required information.
+ ++ +/-
Sticker Stickers are applied to the generic BiB, whereby the stickers
contain all the required information.
- + +/-
Bundle 2 Generic BiBs are bundled, whereby the bundle contains all the
required information.
- ++ ++
Duo-pack 2 Generic BiBs are placed in a box, whereby the box contains all
the required information.
++ ++ +/- / +
For the packaging concepts wrap foil, bundle, and duo pack the packaging material containing the customer specific and legal required information will be removed when the Cafitesse product is used. In these cases the Cafitesse products should still contain the following information; product name, expiration date, manual, and producer information, in order to provide the customer enough information to handle the product correctly. This information has to be provided in the language of the country to which the product is sent to. Therefore, it is important that this information is applied after the freezing process. Otherwise, the generic products kept in stock would increase and the effects of the packaging/labeling postponement strategy would decrease. To apply this information a booklet sticker e.g. could be applied on the generic Cafitesse product after the freezing process. After which it is packed according to the packaging concepts. Table 5.3 gives an overview of some considerations of the packaging concepts.
Packaging concepts Considerations
Sleeve The sleeve does not shrink nicely around the product and the supplier, Fuji Seals,
foresees great technical challenges when this packaging concept is used. As well as problems with applying perforations to open the product.
Wrap foil Requires additional customer specific information on the generic BiB.
Sticker Sticker must be applied on a frozen product. Hence, the right glue for moisture
underground must be used. As well as the generic BiB will deform a bit through the freezing process, this may cause problems when applying the sticker. As well as problems with applying perforations to open the product.
Bundle Requires additional customer specific information on the generic BiB.
Duo-pack Requires additional customer specific information on the generic BiB. Carton box
may cause problems when supplying Japan, since carton package material is not allowed in Japan.
Table 5.3: Consideration packaging concepts
5.3 Conclusion
Based on the findings in this chapter we now can answer the fourth sub-question of this research:
6. INVENTORY
In this chapter the expected average inventory and safety stock are calculated and compared with the current average inventory and safety stock. Furthermore, the expected and current inventory turnover ratios are compared. Thereby answering the fifth sub-question of this research:
• How does the proposed postponement strategy affect the inventories?
6.1 Average Inventory and Safety Stock Calculation
One of the benefits of a postponement strategy is lower inventory levels. It improves the flexibility of inventory through keeping the stock keeping unit (SKU) in generic form at the CODP, which enables safety stock to be reduced whilst offering the full range of finished items (Harrison and Skipworth, 2008). When a packaging/labeling postponement strategy is applied at DE CT&S, the CODP will be moved further downstream between the freezing process and final packaging (figure 5.3). Hence, the inventory of semi-finished product, BiB, will be held there. In order to determine the inventory levels of semi-finished liquid coffee products, first the safety stock level will be determined. The safety stock serves to cushion the impact of forecast errors or deviations in the lead time as well as in the demand during the lead time (Schönsleben, 2007). Different techniques are available to determine the safety stock level based on the nature of the product (table 6.1).
Technique Safety stock Typical use
Fixed Set (manually) quantity New and old items, discontinuous
or lumpy demand patterns, low cost items
Time period Determine by forecasts for future
periods
Critical components, new and old items, discontinuous or lumpy demand patterns
Statistical Calculate via statistical method
based on history
Mature items, continuous or regular demand patterns, deviations in predictable range
One of the techniques is calculating the safety stock based on statistical fluctuations in the lead time. With this technique the safety stock is determined by the demand forecast during the safety lead time. Whereby, the safety lead time is an element of time added to normal lead time to protect against fluctuations. Order release and order completion are planned for earlier dates (before real need dates), according to the time added (Schönsleben, 2007). However, to absorb fluctuations in demand safety lead time is not a sufficient basis to calculate the safety stock level. Therefore, the statistical fluctuation in demand is used as a basis to calculate the safety stock level. Since there are fluctuations in the demand of the liquid coffee products, statistical fluctuations in demand will be used as a basis to calculate the safety stock level for each blend and volume. Appendix 2 gives an overview of the demand patterns and quantity for each blend and volume from week one FY08 up to week 17 FY09 included. Based on those data the necessary figures where determined (appendix 5) to calculate the expected safety stock level, with the formula in figure 6.1, when a packaging/labeling postponement strategy is implemented. Hence, the expected stock is calculated of the generic modules of each blend and volume and compared with the current total stock of the customer specific units of each blend and volume.
Figure 6.1: Formula for safety stock (source: Schönsleben, 2007)
The safety stock is dependent on the service level. The service level is the percentage of order cycles that the firm will go through without stock out, meaning that inventory is sufficient to cover demand (Schönsleben, 2007). For DE CT&S the service level is set at 98% which is agreed with the customer. Without safety stock it is assumed that the demand can be satisfied half of the time (Schönsleben, 2007). If the desired service level is set, the safety stock can be estimated accurately through statistical derivation (Schönsleben, 2007). The service level relates to a safety factor which is the numerical value, a particular multiplier, for the standard deviation of demand (Schönsleben, 2007). Table 6.2 gives an overview of service levels and
safety stock = safety factor * standard deviation of the demand during the lead time where
safety factor = g(service level)
g = inverse function of the integral distribution function chosen and
standard deviation of the demand during the lead time
their corresponding safety factor when demand follows a normal distribution. Based on the service level and the standard deviation of demand during the lead time the safety stock of each blend and volume is calculated. For the blends that follow a normal distribution pattern a safety factor of 2.054 is used, corresponding with a service level of 98% (table 6.2), in the calculations.
Service level Safety factor
50% 0 65% 0.385 80% 0.842 90% 1.282 95% 1.645 98% 2.054 99% 2.326 99.9% 3.090 99.99% 3.720
Table 6.2: Safety factor in relation with service level (Schönsleben, 2007)
For the TQL blends, 605 and 612/613, a lead time of three weeks is used. These three weeks include one week for roasting, extraction, and packaging, one week for freezing and quality inspection, and an extra week for quality inspection. IQ blends 618, 631, 651, 672, and 691, have a lead time of two weeks, since they require one week less quality inspection. Blends 646 and 685 are also IQ blends however they are only produced around once every six weeks. Hence, for these blends a lead time of six weeks is used in the formula. It is assumed all material needed for production is available. Hence, no extra lead time is added for planning and ordering and receiving material. The length of the statistic period used in the formula of figure 6.1 is one week for each blend. Since, the demand pattern is analyzed on weekly demand quantity. Furthermore, the average expected unrestricted use inventory (inventory which is ready for use) level will be calculated with the formula in figure 6.2.
Average inventory = safety stock + order quantity/2 Figure 6.2: Average inventory (Schönsleben, 2007)
two weeks. Blends 646 and 685 are produced once every six weeks. Blend 651 is only produced on order. Hence, the order quantity of blends 605, and 612/613 have to cover one week of demand, blends 618, 631, 672, and 691 two weeks of demand, and blends 646, and 685 six weeks of demand. Consequently, on average the order quantity per blend is the average demand per week times the total weeks of demand it has to cover.
The expected average unrestricted use inventory is calculated with the formulas in figure 6.1 and 6.2 and the figures in appendix 5. The results of the calculated expected safety stock and average unrestricted use inventory will be discussed in the next paragraph and compared with the current safety stock and average unrestricted use inventory level. Blocked stock and stock in quality inspection are not used in the calculation. Since products in quality inspection are seen as work in process. Blocked stock is seen as waste which has to be removed in lean production, which has been deployed at DE CT&S. Therefore, blocked stock and stock in quality inspection are not used in the calculation.
6.1.1 Overview Current and Expected Safety Stock and Average Inventory
Table 6.3 gives an overview of the expected average unrestricted use inventory after implementing a packaging/labeling postponement strategy and its percentage of the current average unrestricted use inventory. The current average unrestricted use inventory level is calculated based on the monthly inventory levels from P03FY08 up to P02FY09 included. Furthermore, the table gives an overview of the expected safety stock level and the current safety stock level. The current safety stock at Diepvriesveem is calculated based on a fixed volume (based on average demand and standard deviation) or fixed days of supply, dependent on the customer.
When analyzing the expected safety stock levels of the 1.25 litre products, a reduction is noticed in blend 605, 612/613, and 631 as expected. Blends 618, 646, and 685 show no reduction, since their demand pattern is discontinuous and cannot be calculated with statistical derivation. Their safety stock is determined by forecasts for future periods, which is currently the basis for determining the safety stock at DE CT&S. Hence, it is assumed that they remain the same. Blend 651 will become an order article. Hence, no safety stock is necessary.
Analyzing the expected average unrestricted use inventory level of the 1.25 litre products. The blends 618, 631, 646, and 685 shows huge increase or decrease.
• Blend 618 shows a huge increase which can be explained by the fact that the demand for blend 618 is increasing rapidly and production needs time to adapt to this increase of demand and bring the current average unrestricted use inventory level to the required level. Hence, the current average unrestricted use inventory level should have been higher. • Blend 631 shows a huge decrease in expected average unrestricted use inventory level and
safety stock. This is due to the fact that the current required stock levels were higher than really necessary. Hence, the current average unrestricted use inventory level and safety stock should have been lower.
• Blend 646 shows an increase of 55%. However, this is such a small blend that it hardly affects the overall result.
• Blend 685 shows a huge increase in average unrestricted use inventory level. This is due to the fact that the current average unrestricted use inventory level is lower than necessary, which can be explained by the fact that the coffee required for blend 685 is hard to get, resulting in stock-out. This lowers the current average unrestricted use inventory level of blend 685.
Overall, the safety stock of the 1.25 litre articles is reduced with 57% and the 2 litre articles with 54%. The average unrestricted use inventory level is reduced with 24% for the 1.25 litre articles and with 38% for the 2 litre articles. Although there are some indications that the current average unrestricted inventory has some bias due to the reasons previous mentioned, which does influence the overall results of the reduction in the inventory. However, due to the fact that the current average inventory is calculated over a period of one year, P03FY08 up to P02FY09 included, the figures gives a good indication of the current average unrestricted use inventory. The savings of these reductions in inventory will be discussed in chapter 8.
6.1.2 Inventory Turnover Ratio
As a result of the decrease in average inventory level, the inventory turnover ratio increases. The inventory turnover is defined is figure 6.3. Hence, if the annual sales remain the same but average inventory level decreases, the inventory turnover ratio increases.
Inventory turnover ratio = annual sales / average inventory level Figure 6.3: Inventory turnover ratio (Levi et al. 2008)
6.2 Inventory Levels Packaging Materials
When the packaging/labeling of the Cafitesse products is postponed the printed broadsheet will be replaced by a generic broadsheet which will be used for all blends. Hence, the demand for the broadsheets can be based on aggregate forecasts which are more accurate, which results in less uncertainty in demand leading to reduction in safety stock (Levi et al., 2008). Furthermore, less safety stock is required for non-specific products compared to the safety stock required for several specific products (Davila and Wouter, 2007).
Based on the stock levels of printed broadsheets at DE CT&S, from week 36, 2007, up to week 52, 2008, with an interval of four weeks (appendices 7 and 8), the current average inventory of printed broadsheet is determined. For the 1.25 litre printed broadsheets the current average inventory is X pieces and for the 2 litre printed broadsheets X pieces (table 6.4).
Current average inventory printed broadsheet 1.25 litre at DE CT&S
Current average inventory printed broadsheet 2 litre at DE CT&S
Total (pc) X X
Table 6.4: Current average inventory broadsheets at DE CT&S
Generic broadsheet 1.25 litre at DE CT&S (pc) Generic broadsheet 2 litre at DE CT&S (pc) Safety stock X X Order quantity X X Average inventory X X
Table 6.5: Expected average inventory broadsheets at Diepvriesveem
6.3 Conclusion
Based on the findings in this chapter we now can answer the fifth sub-question of this research:
7. SERVICE
In this chapter the expected delivery lead time and fill rate with a postponement strategy are discussed. Thereby answering the sixth sub-question of this research:
• How does the proposed postponement strategy affect the service level?
7.1 Delivery Lead Time
The planning at DE CT&S is done in week-buckets. Hence, the lead time is expressed in weeks. Figure 7.1 gives an overview of the lead time. These are the lead times for standard products, products defined in SAP. If a new product is ordered the delivery lead time increases. Since, the graphics of the packaging material have to be approved and the product has to be defined in SAP, which results in additional lead time. Furthermore, when a Cafitesse product, which is made to order, fails quality control the delivery lead time increases as well. The customer has to wait for an extended quality inspection. If the chance for release after extended quality control is lower than 80%, DE CT&S plans a new production (if package materials are available). If the chance for release after extended quality control is higher than 80%, DE CT&S does not plan a new production. However, if the Cafitesse product still fails the quality control new production has to be planned. This increases the delivery lead time significantly.
Figure 7.1: Overview lead time before postponement.
When a packaging/labeling postponement strategy is applied by DE CT&S all products would be kept in a generic state after the freezing process and quality control. The delivery lead time for all products would be two weeks, one week of final packaging, and one week of shipment
Location OpCo Diepvriesveem
DE CT&S
Roasting Extraction Packaging Freezing + Quality ZFIN: Finished product Coffee beans Package materials CODP:
Articles made to order Articles made to stock CODP:
Week 2 Week 3 for IQ blends
Week 3 & 4 for TQL blends
(figure 7.2). The exact days of shipment is dependent on the customer, appendix 1 gives an overview of these exact days per customer. However, this is based on the assumption that the materials for the final packaging are available from stock. Under these conditions the delivery lead time of Cafitesse products made to order is reduced by two weeks for IQ blends and three weeks for TQL blends. Moreover, the products held in a generic state are already passed quality control. Hence, if a product fails quality inspection it does not affect the delivery lead time. The delivery lead time will increase if the material for final packaging still has to be ordered, depending on the delivery lead time of the packaging supplier. The delivery lead time will also increase if the order consists of a new article of which the graphics of the packaging material has to be approved and the product has to be defined in SAP. Furthermore, it should be noted that substantial excess capacity should be provided at the postponed process to enable it to remain responsive when subjected to high demand variability in terms of product mix, and to lesser extent volume. Delivery reliability can suffer when sufficient excess capacity is not provided (Harrison and Skipworth, 2008). Hence, if the capacity of the final packaging process is less than the demand the delivery lead time will increase.
Figure 7.2: Overview lead time after postponement.
Harrison and Skipworth (2008) note that form postponement is preferred to make to stock or make to order when there is a need for greater responsiveness, in terms of shorter delivery lead times than make to order can deliver. Furthermore, Harrison and Skipworth (2008) mention that form postponement is worth implementing if it delivers either reductions in manufacturing costs or improvements in customer service that will provide competitive advantage just like all other operation improvements. Considering this we notice that the delivery lead time for Cafitesse products made to order, 71% of the Cafitesse products and
Location OpCo Diepvriesveem
DE CT&S
Roasting Extraction Packaging Freezing + Quality
Final Packaging Package material &
generic product Coffee beans Package material CODP: All articles
Week 2 Week 3 for IQ blends
