The road to a pan-European distribution network
The location, characteristics and make-or-buy decision of the new European distribution centre(s)
for Steelcase International
Final report
Author: Maarten Res 1064487
Groningen, June 2004
The author is responsible for the contents of this report: the copyright belongs to the author University of Groningen Faculty of Management and Organization Technology Management Groningen, The Netherlands
Supervising professors:
Drs. J.T. van der Vaart Prof. dr. J. Wijngaard Steelcase International
Strasbourg, France
Company supervisor:
N. Goehry-Haessler
Management summary
The objective of this study is twofold and is defined as follows:
o To create a vision for Steelcase International with regard to the establishment of a new European distribution platform: its location, its role, its capacity and the activities that it will perform;
o To provide recommendations regarding the make-or-buy decision and short and long term implementation concerning the European platform.
Steelcase has identified ambitious objectives with regard to distribution performance:
achieve a cost reduction of 20 per cent of the distribution activities, reduction of the lead time from six to two days if the distance is shorter than 1000 kilometres and respond to specific customer demands by offering new distribution services. Therefore, this research focused on a new distribution strategy for the short and long term.
The location and associated cost analysis results’ were that one European platform that is used for all orders that need to be consolidated, is unfeasible. The model that has been created shows that the two platform scenario is more than 26 per cent cheaper, taking into account both warehousing and transportation costs. These distribution centres must be located in the Alsace region and close to Durlangen.
Due to very low customer demand and available space in the factories in Wisches and Durlangen, distribution centre activities are ideally integrated into these factories on the short term to avoid paying for space that is available within the company. Furthermore, a huge work force that is available to perform this work makes that Steelcase possesses the necessary expertise and the labour unions would not be likely to get into new negotiations after the turbulence of the past months. The expected recovering economic situation however, will require more production capacity available, and the space in the mentioned factories could be used for production again in a few years. Another option would be the transfer of the Metal Forming Centre, where mainly semi-finished products are manufactured, to another factory to reduce transportation complexity. Therefore, on the long term an outsourcing strategy with regard to warehousing is recommended.
Three factors underline this statement, which would result in this outsourcing strategy.
The first factor is motorway access, which is very difficult in both Wisches and Durlangen. When distribution activities would be transferred to a third party, a site that is ideally located in the main hubs near the motorway and with good connections to other Steelcase locations and main markets could be selected.
In addition, the warehousing could then be outsourced to a specialised third party who is able to do the necessary investments and obtain economies of scale. This will probably result in better prices and quality, enables Steelcase to focus on the design and manufacturing of products, and creates the flexibility to make use of the warehouses only when needed.
Finally, a more European view on distribution is supported this way which enables more combined transportation, which is requested by many customers, and maybe more efficient transportation as well.
However, Steelcase International should integrate the warehousing activities into the factories on the short term. The first reason for this is that the further development of product platforms would enable a larger product mix per factory. This way, a large reduction of cross-border orders could be achieved, because French orders with Werndl products can then be produced in France for example. Tremendous transportation savings are an obvious result, but service rate would be improved as well, since the customer survey showed that most of the complaints were resulting from ‘international’
orders.
Furthermore, the establishment of the inbound platform(s) create a huge potential for enormous savings when these are combined with platforms for finished products, with economies of scale that can be obtained both on the transportation and on warehousing.
The third aspect is that an IT landscape must be developed first, to obtain effective communication linkages between Steelcase and the third party. This is important to monitor the defined performance and communicate planning changes on a real time basis.
In order to make the final decision on the long term, Steelcase International is recommended the following:
1. Make a realistic estimation of the synergy advantages that can be obtained when it is decided to establish a combined platform. This can only be done when the European inbound flows are mapped and the results of the sourcing from low cost countries are known. In this research, the focus regarding this subject was on the qualitative factors: the choices, conditions, opportunities and risks are discussed into depth.
2. The exact interaction between the two European platforms must be carefully studied according to the product flows at that moment. It is advisable to assign each market to one of the two platforms, but the exact allocation should be carefully studied based on the country’s requirements and logistical situation. This will result in a flexible, tailor- made delivery model per geography.
3. The local distribution platform strategy should also be studied and a benchmark should be performed with regard to warehousing providers on a European scale. This could offer the advantage to have more flexible proximity platforms, so that service level can be guaranteed in the large hubs. Based on this benchmark, the company must decide whether it is advisable to make use of these virtual platforms that are operated by a third party.
4. When the new distribution platform strategy has been defined, decisions must be made regarding the European process management as well. This management of transportation and warehousing can be performed by Steelcase, a third party on a European scale or a
fourth party logistics provider. Each option clearly has advantages and disadvantages, but all possibilities must be taken into account according to the future situation.
Table of contents
Management summary ...2
Table of contents ...5
Preface ...7
1 Company profile...8
1.1 Introduction...8
1.2 Mission ...8
1.3 Description of products...8
1.4 Organisation structure...8
1.5 Steelcase International...9
1.6 Steelcase International strategy...9
2 General problem definition ...11
2.1 Introduction...11
2.2 Problem identification...11
2.3 Problem context...11
2.4 Problem definition...12
2.5 Sub-questions ...12
2.6 Research approach...13
2.7 Conditions and presumptions...14
2.8 Data collection ...14
3 Theoretical framework...15
3.1 Introduction...15
3.2 Quantitative location decisions...15
3.3 Qualitative location decisions ...18
3.4 Warehouse role...19
3.5 Transformation of products on platform ...20
3.6 Warehouse activities...20
3.7 Warehouse capacity ...22
3.8 Warehouse management make-or-buy decision ...22
3.9 Selection of a third-party warehousing provider ...24
3.10 Phase model...24
4 Current general market developments ...25
4.1 Introduction...25
4.2 General logistics developments...25
4.3 General Information & Communications Technology developments ...28
4.4 Conclusion ...30
5 Current situation, future objectives and long term views...31
5.1 Introduction...31
5.2 Inbound process ...31
5.3 Manufacturing process...32
5.4 Outbound process ...33
5.5 Order structure...40
5.6 Distribution cost structure ...41
5.7 Technology ...42
5.8 Conclusions...43
6 Customer demands...45
6.1 Introduction...45
6.2 Survey methodology...45
6.3 Product variants & brands...45
6.4 Distribution & lead time...45
6.5 Value-added services ...46
6.6 Packaging...47
6.7 Conclusions...47
7 Location decision...48
7.1 Introduction...48
7.2 Quantitative research...48
7.3 Qualitative research ...54
7.4 Conclusion ...56
8 Platform characteristics...57
8.1 Introduction...57
8.2 Current platform performance ...57
8.3 Combination of inbound and outbound platform...59
8.4 Distribution & storage strategy ...62
8.5 Transformation activities on platform ...63
8.6 Capacity ...63
8.7 Conclusion ...64
9 Make-or-buy decision...66
9.1 Introduction...66
9.2 Cost analysis...66
9.3 Qualitative outsourcing considerations...67
9.4 Conclusions...68
10 Discussion...69
10.1 Integration...69
10.2 Implications ...71
11 Final conclusion ...73
References...75
Preface
The study that lies before you is the result of a very enjoyable challenge.
I have had the opportunity to do my final research project in Strasbourg at the International headquarters of Steelcase in a fully French-speaking environment. Furthermore, my close contacts with the German organisation showed me the true (cultural) differences between the two countries and the challenges this implies in creating a European view on the distribution process.
In addition, the internship provided me with a tremendous opportunity to get hands on experience in analysing distribution networks, both on the level of real order data and regarding the strategic choices this implies.
I would like to thank Gerhard Bindels for his faith in me and for contacting Steelcase, as well as Manfred Scholz, Alain Dommanget and Philippe Pereira for giving me this opportunity. In addition, a big thank you goes out to Nathalie Goehry-Haessler for helping me find my way in Strasbourg, guiding me on my research process and giving feedback on my assumptions, results and conclusions. Finally, I would like to thank Taco van der Vaart and Jacob Wijngaard for their valuable advise, challenging comments and co- operation to overcome the distance between Strasbourg and Groningen.
Maarten Res
Groningen, June 2004
1 Company profile
1.1 Introduction
In this chapter, an introduction will be given on the organisation of Steelcase Inc, worldwide leader in office furniture. The formulated mission, products and organisation structure will be discussed. This section will be followed by a description of Steelcase International, which is separated from Steelcase North America.
1.2 Mission
Steelcase Inc. is a Fortune 500 company and the world's leading designer and manufacturer of office furniture, with a 2003 revenue of approximately $2.6 billion. Founded in 1912, Steelcase is a global company headquartered in Grand Rapids, Michigan. Approximately 16,000 employees are employed, and there are manufacturing facilities in over 50 locations with a distribution network of independent dealers in more than 900 locations around the world.
The mission is to transform the ways people work by:
o Helping people work more effectively;
o Helping customers use their facilities more flexibly;
o Helping companies use space to attract, retain and inspire workers.
This mission is fulfilled by implementing the strategy, which is built around following central ideas:
o Become a work effectiveness company;
o Achieve enterprise perfection;
o Live the company’s core values;
o Achieve ambitious financial goals.
1.3 Description of products
Steelcase is a global work effectiveness company.
The focus lies on providing knowledge, products and services that enable customers to create work environments that help people in offices work more effectively. Because of this focus, the company has an evolving portfolio of architecture, furniture and technology products and solutions.
Products are offered with a variety of aesthetic options and performance features, and at various price points that address the three core elements of an office environment: architecture, furniture and technology.
Architecture
o Interior architecture. Full and partial height walls with fabric; whiteboard surface materials; raised floors; doors and modular post and beam products.
o Lighting. Task, ambient and accent lighting with energy efficient and user control features.
Furniture
o Panel-based and freestanding furniture systems. Moveable and reconfigurable furniture components used to create individual workstations and complete work environments. Systems furniture provides visual and acoustical privacy; accommodates power and data cabling; and supports technology and other work tools.
o Storage. Lateral and vertical files, cabinets, bins and shelves, carts, file pedestals, and towers.
o Seating. High-performance, ergonomic, executive, guest, lounge, team, health care, stackable and general use chairs.
o Tables. Meeting, personal, learning and coffee tables.
o Textiles and surface materials. Seating upholstery, panel fabric and wall coverings.
o Desks and Suites. Wood and non-wood desks, credenzas and case goods.
o Work tools. Computer support, technology management, organisers, information management products and portable whiteboards.
Technology
o Infrastructure. Infrastructure products, such as modular communications, data and power cabling.
o Appliances. Group communication tools, such as interactive whiteboards, image capturing devices and web based interactive signs and space scheduling systems.
1.4 Organisation structure
Steelcase Inc. consists of two geographically separated divisions: North America and International.
Steelcase Inc.
Steelcase North America
Steelcase International
Figure 1-I Organisation structure Steelcase Steelcase North America is responsible for the United States and Canada. The division has a turnover of 1.9 billion Euros, employs 11,500 people and runs 31 factories. It is market leader with a market share of 18.9 per cent and has working relationships with 650 dealers. The rest of the research will focus on the other division, Steelcase International.
1.5 Steelcase International
Steelcase International has been established mainly through acquisitions. Since 1974, Steelcase Inc. has acquired many companies all over the world. All operations except for the region North America are part of the division Steelcase International, which is headquartered in Strasbourg, France.
Total turnover is 700 million Euros with its 5,000 employees, while the total number of factories is 14. The division has an 8.1 per cent market share and is co-operating with 350 dealers.
Steelcase International consists of six geographies (see Figure 1-II). This research is only concerned with the European subsidiaries, in light turquoise: France, Germany, Spain and Italy.
Steelcase Inc
North America International
France Germany Spain Italy Morocco EMEA
Figure 1-II Organisation structure Steelcase International
1.6 Steelcase International strategy
The strategy of the division is built around four axes:
Customer satisfaction
The company wants to optimise the operational processes to satisfy the customer. This satisfaction of the clients depends on three issues. The first aspect is a quality and service performance which is superior to the competition. These items need to be improved by clear and simple manufacturing processes.
Furthermore, Steelcase International needs to be responsive to a clearly international demand.
Global customers need to be able to buy the same Steelcase product around the world, with the same service level and product quality.
On the other side, it is needed to respond to the local customer needs as well. In order to achieve this, the company must be near the local client and develop local competencies, so that the client’s needs and culture can be taken into account.
Transforming the way people work
The enterprise wishes to position itself as a performance delivering company. This means that it will focus on aesthetics as well as a high technology level and a wide range of services, so that an added innovative value can be offered to the customer instead of only the delivery of an office furniture product.
In the United States, Steelcase Inc. has acquired an architecture of buildings. Steelcase could propose to construct a building while integrating the technology and ideal workspace usage. At the headquarters, this centre has been established and the wish is to create a new social chair based on this expertise in France.
Achievement of ambitious financial results: EVA=0 To be able to give the investors in the company a suitable return, revenue of 12 per cent needs to be achieved after taxes, which corresponds to an Economic Value Added (EVA) of 0. Currently, between the acquisition of Steelcase Strafor and the necessary investments for the future development of the company, this return-on- investment is negative (less than 12 per cent).
This objective has to be achieved within three to four years, while continuing to invest.
Living the Steelcase values
The Steelcase company wishes to express its values to all its employees over the world. These are:
o Act with integrity;
o Tell the truth;
o Keep commitments;
o Treat people with dignity and respect;
o Promote positive relationships;
o Protect the environment;
o Excel.
2 General problem definition
2.1 Introduction
In this chapter, the reasons for this research will be described. These are recent changes in the European distribution strategy, which preceded the research process. In the problem identification, the objectives with regard to the physical distribution will be described briefly.
The problem context will identify the current European distribution network and other projects that will have an impact on this research.
This is followed by the problem definition, with the objective and main question of the research, and the sub-questions. The chapter ends with the research approach, conditions and presumptions and ways of data collection.
2.2 Problem identification
Logistics need to become a competitive advantage for Steelcase International. It is essential that products arrive at the customer on time, complete and damage free. The company has formulated three main objectives with regard to its distribution strategy to be reached in the near future. The first goal is to achieve a cost reduction of 20 per cent of the distribution activities. This must be achieved by improving efficiency. In addition, the lead-time can be improved. Transportation lead-time is to be reduced by 60 per cent. Currently, it takes six days from the end of fabrication until delivery.
The goal is to serve the client within two days if the distance is shorter than 1000 kilometres.
Finally, by offering new distribution services, Steelcase International wants to be able to respond to specific customer demands.
One of the tools to achieve this is by establishing a new European distribution centre to synchronise product flows in Europe. In first instance, this concerns an outbound platform that is used for physical distribution purposes.
Orders that are too small (in practice smaller than 20 cubic metres) to deliver directly to the customer from the manufacturing plant or need to be regrouped, will pass by the platform before being delivered to the dealer or customer. A logistics service provider could be responsible for the warehousing and transport. This research will provide recommendations on the location and characteristics of this new European
distribution platform as well as the make-or-buy decision with regard to the warehousing.
2.3 Problem context
In this section, the environment is presented in which the research is conducted. For example, the set-up and objectives of other projects that will have an impact on the distribution structure are discussed.
Current European factories and distribution centres In the current situation in the Alsace region, there are production facilities in Sarrebourg (chairs), Marlenheim (desks) and Wisches (storage). The Strasbourg site is closing down and served as storages factory, Metal Forming Centre (to be moved to Rosheim) and distribution centre (Cédis). Another plant in Western France is located in Senonches. In Germany, there are production sites in Rosenheim and Durlangen, while Madrid is the manufacturing location in Spain and Bologna for Italy.
In Waltham, near London, and Ile de France (Paris), local platforms can be found. Some of the factories perform platform activities as well.
In Figure 2-I and 2-II of the appendix, an overview can be found of the exact location and product mix per factory.
Strasbourg site closure
The Strasbourg factory, where storage products were manufactured, will close halfway 2004. The production from Strasbourg will be added to the remaining factories. The small pedestals will move to Wisches, while the large storages will be transferred to Marlenheim. The Metal Forming Centre will move to Rosheim. The distribution centre will close down in February 2005, which activities will be transferred to the platforms at Wisches and Marlenheim for the short term. In this study, the role of a new European distribution centre will be redefined.
Distribution strategy
There are great differences in physical distribution planning and strategy between the different countries, which have a great influence on the distribution structure. This will be discussed in depth in the analysis of the current situation in chapter 5. In France, with the movement of production from Strasbourg to Wisches, Marlenheim and Rosheim, a multi- picking strategy will be implemented as well.
This means the trucks can pick up finished products at different factories before going to the client or dealer. This is also known as milk run.
5&2 project
Only 5 per cent of the time a product is in the possession of Steelcase International, value adding activities are performed to the product.
To streamline the product flow, the 5&2 project will be implemented. This means that the time to complete and deliver a customer order will only take seven working days: five days for procurement and manufacturing and two days for delivery within 1000 kilometres. In order to achieve these targets, total time on the new distribution platform may not exceed four hours, which will require a great change in platform role and performance compared to the current situation. Communicated delivery time to the customer will be ten days.
Low cost country sourcing
To reduce the costs of raw materials, components or tooling will be purchased in low cost countries without reducing quality of products and customer service. This is done with a long term perspective and taking into account long term trends in labour costs, political risks but also the short term and long term consequences on the international industrial system. The first shipments from Asia have been received, with a target to source out of Asia at least 20 million Euro per year in the Fiscal Year 2006. This should result in a three million Euro saving per year. This means more raw material stocks are required, which will be stored at a new inbound platform. A temporary platform solution for the first eighteen months has been implemented. This means the appropriate supply chain to sustain this strategy must be defined and implemented, which could include combining this inbound distribution centre with the new (outbound) distribution platform.
Streamlining of Information Technology
To assist the streamlining of the product and information flows, an SAP Sales & Distribution system is used. Under the project name of ELIS, the goal is to implement this software all over Europe. ELIS stands for European Logistics Integration System and has been developed for Steelcase International. This enables the company to control the European information flows and the organisation of processes. It
improves operating performance by offering real-time information that is standardised and shared, a diverse and international workforce, complete product internal tracking for greater security on deliverable items, increased quality of service and delivery, a one customer, one order strategy, and support best practices and continuous improvement strategy. Only Germany, that uses software called MBI, does not use SAP yet.
Zero-packaging strategy
To reduce packaging costs, the organisation wishes to use as little packaging as possible. For direct deliveries, the possibility is investigated to transport products without any packaging at all.
Products are transported in trolleys to protect against damage. Advantages are reduced packaging cost and environmentally friendly packaging. This has to be taken into account with the design of the new platform and the calculation of the necessary capacity.
2.4 Problem definition
The objective and main question are defined in the following section.
Objective
The objective of the study is twofold:
- To create a vision for Steelcase International with regard to the establishment of a new European distribution platform: its location, its role, its capacity and the activities that it will perform;
- To provide recommendations regarding the make-or buy decision and short and long term implementation concerning the European platform.
Main question
“What strategy can be defined for the location and characteristics of the European distribution platform and how should the warehouse be operated according to Steelcase International’s needs and objectives?”
2.5 Sub-questions
To be able to answer the main question, several sub-questions are formulated. To analyse the platform needs, first an analysis of the current situation is provided to get further insight in the latest market developments, the way the supply chain process is organised, the influence of ICT with regard to distribution and customer
expectations. This is necessary to understand its complexity and main issues.
This will be followed by an analysis of the optimum location and other characteristics of the new distribution centre. These factors together result in a re-design of the distribution structure that is in line with the first part of the main question.
The outsourcing decision of the warehousing activities will give further insight in the costs of the different scenarios, after which the final decision can be made. This corresponds to the second part of the main question and will be in line with the stated objectives of cost reduction, lead time improvement and offering value-added services.
The sub-questions are formulated as follows:
1. How can the current distribution structure be described?
a. What are the main distribution developments on the general market?
b. What does the supply chain look like?
c. How is the order structure built up?
d. How can the distribution cost structure be described?
e. What is the influence of Information &
Communication Technology on the distribution process?
2. What do customers expect from the distribution process at Steelcase International?
3. Where should the distribution platform be located?
4. What should be the role of the platform?
5. Which activities should be performed within the platform?
6. What should be the capacity of the platform?
7. Do the activities need to be outsourced or performed by Steelcase International?
2.6 Research approach
The following steps will be performed during the research phase, which will lead to an answer to the main question. The research will consist of three phases:
o Analysis of the current situation and customer requirements, which will influence Steelcase International’s distribution strategy (step 1 and 2);
o Definition of distribution platform location and characteristics (steps 3-6);
o Warehousing make-or-buy decision (step 7).
The report is built up as follows:
6. Platform capacity 7. Make-or-buy decision 1. Analysis current situation
3. Location decision 4. Platform role 5. Platform activities 2. Customer demands
Figure 2-I Research steps
In chapter 3, an extensive phase model will be presented which is the base for the steps that are presented here. This is based on the relevant literature that is linked to the sub-questions.
In chapter 4, sub-question 1 (a) will be answered which discusses the latest market developments with regard to logistics and ICT. It is important to keep these developments in mind when analysing the internal situation.
The subject of chapter 5 is the current organisation of the relevant business processes within Steelcase International as mentioned in the sub-questions 1b-e. This chapter provides further insight into these processes before the platform characteristics are determined.
Chapter 6 will present the results of a survey that was conducted to evaluate customer needs and demands (sub-question 2). These also serve as input to the new distribution strategy in general and warehouse strategy in particular.
Sub-question 3, which is concerned with the location analysis, can be found in chapter 7 and is based on the conclusions of the previous chapters and the location model that has been designed.
Chapter 8 will deal with the other platform characteristics: the role, activities and capacity (sub-questions 4-6). This will give further insight in what the warehouse will look like and for which flows it should be used and will be discussed after the location conclusions, because the location is very important for the further platform characteristics.
After the conclusions about the platform characteristics, the make-or-buy decision will be made in chapter 9 (sub-question 7). The
question whether the warehousing activities should be outsourced or not will be answered in this chapter.
This is followed by a discussion that deals with the integration between the different subjects (chapter 10), and the final conclusion (chapter 11), in which the main question will be answered.
2.7 Conditions and presumptions o A restriction to the research will be that only the division Steelcase International will be part of the research. Within this division, which is responsible for the world except North America, the research will be based upon the European situation. The interaction with other continents will be taken into account only with respect to the development of Asian sourcing, the project of searching for competitive suppliers in Asia;
o The analysis of the developments currently taking place on the market will include the issues that are relevant for creating a vision for the new European distribution platform;
o The research will consider the establishment of one or two European distribution platforms, taking into account the capabilities and possible roles of local platforms;
o Processes within the production plants are left out of the research as far as they are not concerned with physical distribution;
o The objective in the location research is to minimise the total distribution and warehousing costs, taking into account the volumes of the product flows and other objectives;
o The research will be conducted following the guidelines set by the Faculty of Management and Organization of the University of Groningen;
o The field research will take until the end of May;
o The final report will be finished at the end of June.
2.8 Data collection
In the study, different ways of obtaining data will be used. Three ways of data collection exist:
usage of existing information, obtaining of data through observation and collection of data through written or oral interviews.i In this study, the data will mostly be obtained through the
usage of existing information and oral interviews.
Usage of existing information
Since the change process has already started, a lot of information is already available within the organisation of Steelcase International.
Furthermore, analyses are based upon the order history.
Oral interviews
Other information that is necessary for the data collection process will be obtained by interviews with responsible employees concerning a certain research area. The customer demands will be investigated with a questionnaire that has been designed and interviews are conducted with a selection of dealers by phone.
3 Theoretical framework
3.1 Introduction
Now that the objectives and research approach have been clarified in chapter 2, a theoretical framework can be presented with the relevant research that has been done already. In this chapter, an extensive literature overview is provided with regard to all the steps presented that are necessary to answer the main question:
location decisions, warehouse role, activities and capacity and the outsourcing decision. At the end of this chapter, this will be concluded in the phase model, in which the coherence between the relevant aspects, the research approach and the outline of the thesis has been visualised. This way, each paragraph is concerned with one of the sub-questions of the previous chapter.
Location decisions are seldom based only on quantitative or qualitative research. The combination of these two provides the best input in the final site selection. First, an overview of quantitative techniques will be provided, followed by useful qualitative aspects that need to be taken into account.
3.2 Quantitative location decisions The sequence of steps that has been found effective according to Bauer (2002) with regard to network configuration for distribution problems is as shown in Figure 3-I.
Her conclusions with regard to Distribution Network Planning are threefold.
First, the largest obstacle is often the preparation of a satisfactory database.
A second observation is that “not only cost savings are an important objective when modelling network optimisations. The transparency of the system and processes to the client, as well as the ability to illustrate its internal logic to the practitioners is most critical”.ii
The third conclusion is that heuristics have to be used (designed and tested) when the data provision by the client is not satisfactory.
STEP 1: INDENTIFYING PROBLEM SITUATION Establishment of a project plan
STEP 2: DATA ANALYSIS Clients’ order or shipment transaction data Testing of data with regard to completeness and plausibility
Correction of gaps and errors
STEP 3: INITIAL CALCULATION Establish a baseline to compare scenarios
Estimation of different scenarios
Verification of gap between status quo and client’s perception
STEP 4: SCENARIO CALCULATION & COMPARISON Based on several network configurations
STEP 5: SOLUTION
Interactive and incremental step by step process Satisfaction of optimisation targets
Figure 3-I Steps quantitative location research
Location decisions in general focus on selecting the number and location of distribution centres.
This is a very complex decision making process, since the number of locations, the alternative location sites and the stocking strategies for each location have to be considered. Data intensity is required because the analysis requires detailed demand and transportation information.
There is a wide rage of facility location problems, which depends upon several factors.
It could concern one single or multiple facilities and one or multiple different objectives. In addition, different problem types exist (planar, network, discrete, etc.), as well as different types of distance function. Finally, the problem could involve special assumptions and restrictions.iii As this research will focus on a specific type of problem, only this theory will be presented here.
It concerns the following problem type:
o One or two facilities (distribution platforms), since the rest of the plants and distribution centres are part of the context;
o One single objective: minimising the total distribution costs, taking into account the volumes of the product flows (time is assumed to be directly related here to distance) and the costs of operating distribution centres;
o The problem type is planar, which means a continuous area without a pre-defined set of possible location alternatives;
o The type of distance function used is a general d-type (straight line);
o No special assumptions or restrictions are involved.
The tools used to support these kinds of location analysis generally fall into the categories of analytic techniques, optimisation or linear programming techniques, and simulation. Each is described and discussed, after which the final choice is made for one of the techniques. The exact way of calculating is described in the appendix of chapter 7.
Centre of gravity calculations using analytic techniques
This method is appropriate for locating a single distribution centre or plant. A number of methods, both mathematical and nonmathematical, can be applied to the problem of a single location. The cost and complexity of the technique should be matched to the difficulty of the problem. The model is based on Cartesian co-ordinates, where the horizontal or east-west axis is labelled the x-axis and the vertical or north-south axis is labelled the y-axis.
The co-ordinates are overlaid on the relevant geography.
The goal of this technique is to minimise transportation costs. Generally, these are a function of time, weight and distance. To reduce complexity, sometimes not all of these factors are considered in the final analysis. Four solution methods exist: ton-centre, mile-centre, ton-mile- centre and time-ton-mile-centre solution. As the titles suggest, the first three are limited to variables related to weight and distance. The fourth includes both weight and distance plus time as cost determinants. The weighted average and median location of the x and y co-ordinates can be used in practice.
In literature, different visions on the exact ways of calculating exist. The calculations of the ton- centre solution by Bowersox et al. (1996)iv is the same as the one presented by Coyle et al. (1996)v as the ton-mile-centre solution. Bowersox indicates that for the formula used, the location point represents the centre of gravity and therefore only weight is given consideration while time and distance are neglected. Coyle claims the distance is in fact included, by first moving horizontally and then vertically instead of in a straight line. In reality, the centre of gravity does not minimise total distance, which can be illustrated by a simple example. When a situation with only two destinations is considered, with twice as much demand in point A as in point B, the centre of gravity will be located in between A and B with the distance to A being half the
distance to B. However, to minimise travelled distance, the optimum can be found in point A as every move closer to B will increase total distance. If there is one point that represents half the demand, the optimum location will always be located in that point, no matter where the other points are located. The method of minimising real distance travelled is superior, which requires an iterative process with several iterations until a solution is found within an acceptable range.
Coyle notifies that transportation rates can also vary between different destinations and for different commodities. For example, transportation between factory and platform may be cheaper than from the platform to the final customer, because the transport can be optimised in a more efficient way. He also recognises that rates are ‘tapering’ in real life, which means the transportation costs will decrease per kilometre as distance is increased.
In this research, three different calculation methods will be used. Because kilometres and cubic metres are used instead of miles and tons, these methods will be named distance-centre, volume-centre and volume-distance-centre.
Bowersox’ definitions and methods of calculation will be used because these deliver superior volume-distance-centre calculations.
Coyle’s volume-distance-centre solution will also be used as it corresponds to Bowersox’ volume- centre solution. The influence of time will be considered as directly dependent on distance, because duration of different tours are highly fluctuating as well. The influence of fluctuating transportation rates is recognised by investigating different scenarios. Utilisation of real transportation rates proved to be impossible, because of the above mentioned tapering rate system, usage of different carriers to different destinations, changing rates and changing carrier choice over time (for example the contract with one of Germany’s key transportation company will be ended in the coming months).
Location-allocation decisions using linear programming
When designing distribution networks, the centre of gravity is sometimes not the optimal solution. When more than one distribution centre is involved, the products have to be assigned to a warehouse, which creates a location-allocation problem. This needs to be
optimised simultaneously, because sub- optimisation can lead to poor location decisions.
The situation becomes even more complex when the routing of products is also considered, in which case we speak of location-routing problems.
Analytic techniques are not sufficient any more and therefore heuristics have to be used in these cases. Linear programming methods, which are classified as an optimisation technique, are one of the most widely used strategic and tactical logistics planning tools. Linear programming selects the optimal course of action from a number of available options while considering specific constraints.
An early definition of optimisation regarding logistics is “[a consideration of] the aggregate set of requirements from the customers, the aggregate set of production possibilities for the producers, the potential intermediary points, the transportation alternatives and develops the optimal system. The model determines an aggregate flow basis where the warehouses should be, where the stocking points should be, how big the warehouses should be and what kinds of transportation options should be implemented.”vi
In order to solve a problem using linear programming, several conditions must be satisfied. First, two or more activities or locations must be competing for limited resources. Second, all pertinent relationships in the problem structure must be deterministic and capable of linear approximation. Unless these enabling conditions are satisfied, a solution derived from linear programming, while mathematically optimal, may not be valid for logistical planning.
Within optimisation, distribution analysts have used two different solution methodologies for logistics analysis. The first methodology is network optimisation, which treats the distribution channel as a network of nodes (distribution centres) and arcs (transportation links). Costs are incurred for handling goods at nodes and for moving goods between nodes.
Rapid solution times, ease of communication between specialists and non-specialists and the incorporation of fixed costs to replicate ownership are the primary advantages of network optimisation. Traditional disadvantages are the size of the problem that can be solved and the exclusion of fixed-cost components,
which limits the capability to optimise both fixed and variable costs for production and distribution facilities.
The second methodology, mixed-integer programming, reduces this problem using a combination of network organisation and mixed-integer programming. Its primary advantage is that fixed as well as different levels of variable costs can be included in the analysis.
This means that also non-integer values can be taken into account.
The constraints on problem sizes that can be solved are reduced by the application of decomposition to the solution techniques, which provides a procedure for dividing multi- commodity situations into a series of single- commodity problems. This procedure has been incorporated a multi-commodity distribution system design program which follows a two- stage iterative process. First, given the combination of possible distribution centres, individual commodities are optimally assigned to the centre that minimises cost in a manner similar to basic transportation solutions.
However, the procedure is formatted on a multi- product basis using the decomposition technique. Second, a mixed-integer algorithm is utilised to enumerate the facility structure in terms of individual commodity customer’s assignments. The combined solution is tested for optimality, and the two-step procedure is repeated until convergence is within accepted tolerance.
Linear programming confronts some major problems when dealing with complex logistical system designs. First, to format a comprehensive design, it is necessary to develop explicit functional relationships for the full range of design options. The functional relationship must consider all possible combinations for suppliers, production locations, distribution locations, wholesalers, markets and products. The sheer number of alternatives and the associated constraints result in a very large problem.
Second, the optimality feature of the technique is only as valid as the design definition. Third, the capability of existing linear programming procedures is limited by the number of echelons or stages in the distribution system and by the problem size.
Simulation techniques
Simulation can be defined as “the process of designing a model of a real system and conducting experiments with this model for the purpose of either understanding system behaviour or of evaluating various strategies within the limits imposed by a criterion or set of criteria for the operation of a system”.vii
Static simulation replicates the product flow as if it all occurred at a single point during the year.
The major expense components are inbound freight (plant to distribution centre), fixed and variable distribution centre cost, customer freight (distribution centre to the customer) and inventory carrying cost. Dynamic situation on the other hand, evaluates system performance across time.viii
Static simulation identifies the ‘best’ solution by comparing the total cost and threshold service capabilities of the combinations resulting from a deletion procedure. The cost and service characteristics of alternative networks are compared, but there is no guarantee that the result will be the optimum or even near the optimum configuration.
“The main advantage of static simulation is that it is simpler, less expensive to operate, and more flexible than most optimisation techniques. […]
Unlike mathematical programming approaches, simulation does not guarantee an optimal solution.”ix Furthermore, all the exact necessary cost parameters must be known and very exact, to be able to serve as input into the model.
Technique selection
Linear programming and simulation techniques can be adequate tools to optimise distribution networks. However, it is outside the scope of this research to redesign the complete distribution strategy. The individual order and planning structure and the interaction with local platforms, as well as the factory mix per order could not be incorporated into a single model.
Furthermore, a constraint to the linear programming method is that multiple locations must compete for limited resources. This is not the case for a single platform scenario. For simulation techniques, a proposed solution can be far from optimal.
Finally, the cost structure that serves as input to the model must be very accurate and reliable to avoid mistakes in decision making.
By making use of different scenarios, the sensitivity of the centre of gravity methods can be analysed which has proven to be sufficient to design an effective location model. This analytic technique is appropriate for localisation of one or two distribution centres. The transportation costs are assumed to be minimised by minimising total distance travelled. In chapter 7, with the location analysis as subject, this will be discussed in detail.
Data requirements
First, a definition of the market is required to obtain a classification of demand assigned to a geographic area. The selection of a market definition method is extremely important, by using county, standard metropolitan statistical area (SMSA) or postal codes. A product definition is usually not necessary, but a network definition is very important. Market demand is needed to define shipment volume to each geographic area identified as a market. Inbound and outbound transportation rates are also a major data requirement for location analysis, so these must be provided for shipments between existing and potential distribution channel members and markets, as well as for each shipment size and for each transportation link between distribution centres and markets.
Finally, the variable and fixed costs associated with operating distribution facilities are needed, such as labour, energy, utilities, materials, facilities, equipment and supervisory management. x
3.3 Qualitative location decisions Qualitative factors that are relevant for the final location decision are not very thoroughly described in scientific literature. These aspects are more based on feeling: ‘soft’ rather than
‘hard’ factors. Several of them are discussed.
First, the method of transportation that is used, is essential for the definitive choice. Whether transportation by truck, train, boat or aeroplane is used, the necessary infrastructure needs to be available. In addition, the use of multi-model transport could increase, with reduced cost, increasing driving times and lower environmental impacts.
