DELIVERY RELIABILITY
PERFORMANCE
Delivery reliability performance of
De Beer Lakfabrieken B.V.
A research to improve the delivery reliability performance of
a car refinish manufacturer
Author: Bertus Meun Student number: 1483064
University of Groningen
Faculty of Economics & Faculty of Management and Organization
MSc Business Administration, specialization Operations & Supply Chains
Supervisors De Beer: M. ten Berge J. van Geresteyn Supervisor RUG: Dr. J.T. van der Vaart Co-assessor RUG: Dr. J. de Vries
MANAGEMENT SUMMARY
De Beer Lakfabrieken B.V. has been developing, producing and selling varnish, primers, hardeners and thinners since 1910. The products are exported to more than 90 countries all over the world. De Beer is a continuously growing car refinish manufacturer. The company has ambitious long term goals, including an expansion of its market share from 1.5% to at least 5% of the global car refinish market within the next 10 years. This growth will be at the expense of competitors and therefore De Beer has to outperform them to win customers. This puts a lot of strain on the relatively small organization and that is why De Beer is searching for ways to improve its internal organization as well as its overall performance.
This research intends to help De Beer in this endeavour by researching the delivery reliability performance of the company and, if possible, by giving suggestions to improve it. The delivery reliability performance is divided in three components: on time delivery, complete delivery, and delivery of the right products. The management of De Beer receives a document in which is described what the situation with regard to the delivery reliability performance was in 2005, which factors caused this performance, and if and how these factors can be improved. Next to this they receive the database used for the analysis.
The following main research question gave direction to the research project:
“What are the possibilities to improve the delivery reliability performance of De Beer towards its customers?”
To answer this question several sub questions were asked of which the most important deal with the delivery reliability performance, its causes and ways to improve the performance.
What is the current delivery reliability performance, measured over 2005?
On time delivery analysis shows that 19038 of the 91729 order lines received in 2005 were late, which corresponds to 20.75%. A striking fact is that 51% of all late order lines was only one day late and that only 9.2% of all late order lines was more than a workweek late. Although the company distinguishes twelve different customer regions and 51 different product groups, a small part of these account for a large part of the total number of late order lines. Only four regions account for almost 70% of all late order lines and only 6 product groups account for more than 70% of all late order lines. The regions are S02, S03, S05 and S06, and the product groups are 103000, 104000, 229000, 272000, 273000 and 275000.
ordered and amount delivered. Relative to the total amounts involved this was a very slight difference. An important cause for this difference is probably that products usually can only be ordered in a fixed amount. Also, if products cannot be delivered the customer sometimes decides not to wait for them or even cancel them. Further analysis showed by means of a complaints analysis that De Beer received 25 complaints about receiving too many products and 123 complaints about receiving too few products. Compared to the more than 5500 orders received this is not a very large number of complaints. Also, the analysis showed that a very small number of order lines and ordered amount was considered a backorder. Furthermore, only a part of these backorders was actually delivered late.
Right delivery analysis showed by means of a complaints analysis that 44 complaints were identified about receiving the wrong product. In comparison to the more than 5500 orders that were received this is a fairly small amount.
What are the causes of the current delivery reliability performance?
# On time delivery Complete delivery Right delivery
1 Order entry error Ordering error by customers Ordering error by customers
2 Inspection times Order picking error Order picking error
3 Planning (average processing times) Production error Production error 4 Extra-operational causes:
- Priorities Sales department - Insufficient communication
Order entry error or mutation error
Order entry error or mutation error
In what way can these causes be dealt with in order to improve the delivery reliability performance?
Several possible solutions are discussed to improve the current delivery reliability performance. These solutions result in a list of recommendations, which can be found in chapter 8.
# Solutions
1 Implementation of forecasts to anticipate demand.
2 Adjustment of inventory policy for finished goods to reduce stock outs.
3 Adjustment of planning technique to deal with mixed-mode production and incorporation of forecasts in the long term planning.
4 Incorporation of QC&I adjustments to batches more regularly in the planning.
5 Instruction of employees about the order entry errors and the interdependencies of the dates. 6 Checking entered orders periodically for inconsistencies.
7 Instruction of employees to check their work before they proceed.
8 Improvement of communication between departments, especially between Planning, QC&I and the shop floor.
PREFACE
This research is my master thesis for the Master of Science Business Administration, specialization Operations & Supply Chains, at the University of Groningen. The context of the research is a growing medium-sized company with ambitious long-term goals: De Beer Lakfabrieken B.V., located in Lelystad. Because of these goals the company wants to improve its performance. This research focuses on the delivery reliability performance of De Beer Lakfabrieken B.V. during 2005.
At this point I would like to thank all the employees at De Beer Lakfabrieken B.V. for all the help they provided and for making my time at De Beer as pleasant as it was. Without their help and willingness to answer all my questions this research report would not have been possible. Specifically, I would like to thank my supervisors at De Beer: Martin ten Berge and Jacco van Geresteyn. Their involvement, guidance and enthusiasm was indispensable for this research and my motivation.
Also, I would like to thank my supervisors from the University of Groningen: Taco van der Vaart and Jan de Vries. Their insights, comments and patience during this research were invaluable for the end result.
TABLE OF CONTENTS
MANAGEMENT SUMMARY ... 3
PREFACE ... 5
TABLE OF CONTENTS ... 6
-PART ONE: INTRODUCTION 1) DE BEER LAKFABRIEKEN B.V. ... 9 -1.1 History ... 9 -1.2 The organization ... 10 -1.3 The market ... 11 -1.4 The products ... 12 2) RESEARCH DESIGN ... 15 -2.1 Initial motive ... 15 -2.2 Problem statement... 16 -2.3 Research method ... 17 -2.4 Sub questions... 18 -2.5 Methodology... 18
-PART TWO: ANALYSIS 3) THE ORDER CYCLE ... 22
-3.1 Activities in the order cycle ... 22
-3.2 Possible order cycles ... 28
4) THE ORDER PORTFOLIO ... 31
-4.1 The data files ... 31
-4.2 General characteristics of the order portfolio ... 32
-4.3 Most ordered product groups and products... 34
-4.4 Impact of Dutch customers ... 35
-4.5 Standard processing time... 35
5) DELIVERY RELIABILITY PERFORMANCE ... 38
-5.2 On time delivery ... 39
-5.3 Complete delivery ... 45
-5.4 Right delivery ... 48
6) POSSIBLE CAUSES ... 49
-6.1 General remarks ... 49
-6.2 Structure of the analysis ... 50
-6.3 On time delivery ... 50
-6.4 Complete delivery ... 60
-6.5 Right delivery ... 60
-6.6 Overview of identified causes ... 60
-PART THREE: REDESIGN 7) POSSIBLE SOLUTIONS... 63
-7.1 On time delivery: operational solutions ... 63
-7.2 On time delivery: administrative solution ... 68
-7.3 Complete delivery ... 69
-7.4 Right delivery ... 69
-7.5 Overview of solutions ... 70
8) CONCLUSIONS AND RECOMMENDATIONS ... 71
-8.1 General conclusions... 71
-8.2 Recommendations... 73
-8.3 Further research ... 74
-PART 1
INTRODUCTION
DE BEER LAKFABRIEKEN B.V.
De Beer Lakfabrieken B.V. has been developing, producing and selling varnish, primers, hardeners and thinners since 1910. The products are exported to more than 90 countries all over the world. In the Netherlands the company has two facilities, one in Lelystad and the other one in Beuningen, and it also has subsidiaries in the United Kingdom and Australia. The facility in Beuningen, however, is being closed and relocated to the facility in Lelystad during this research. In the Netherlands approximately 135 employees are employed. De Beer is one of the few Dutch varnish producers that is specialized in producing car repair varnish.
This first chapter provides some additional information about De Beer so the reader will know what kind of company it is. First, the history of the company will be briefly outlined, starting from its founding till today. This is followed by a description of how the company is organized. After that, the market on which the company is active will be discussed, including the competitors and the customers. Finally, the different products the company produces and sells to its customers will be discussed.
1.1
History
De Beer Lakfabrieken B.V. was founded in 1910 as a trading company in paint, varnish and related products. These commodities were mainly imported from Germany. In the 1950s the company started producing paint and varnish itself. The trading company was located in Venlo, while the production facility was located in Voorschoten. During the 1980s the company started exporting its products, primarily to the Middle East.
In 1993 De Beer was taken over by one of its customers, MoTip B.V. This company was looking for ways to diversify their activities. In January 1998 MoTip B.V. merged with DPCC, which was composed of several companies in the chemical sector. All the involved companies became part of the MoTip Group N.V., which develops, produces and sells paint and repair and maintenance products.
After several years De Beer left the MoTip Group N.V. and became independent again for approximately two years. In January 2004 the company was taken over by The Valspar Corporation. Valspar is headquartered in Minneapolis in the United States and is one of the largest global coatings manufacturers in the world. Valspar has more than 7000 employees in over 80 locations around the world.
1.2
The organization
The organizational structure of a company shows in what way tasks and responsibilities are divided into distinct groupings, and how the responsibility and coordination relationships between the groupings are defined (Slack and Lewis, 2002). The organizational structure of De Beer is changing relatively often due to new insights and experiences. It is primarily organized according to an U-form, which gives prominence to functional groupings of resources (Slack and Lewis, 2002). The organization is divided into the functional groupings Finance, Human Resources, Sales and Marketing, Operations, Technology, and Projects. These functional groupings are further subdivided into departments. Within these departments different grouping criteria are applied, e.g. geographical area’s in the Sales Office, and clustering similar technology in Production & Logistics. The organization chart is shown in figure 1. Although there is some shifting in responsibilities and staffing between departments the overall picture in figure 1 still applies.
Some of the abbreviations in figure 1 need some further explanation:
HSE (currently HEQS): Health, Environment, Quality and Safety. This department deals with all issues regarding health, environment, quality and safety.
QC&I: Quality, Control & Improvement. This department safeguards the established product quality levels of the department Production & Logistics.
TIC: Technical Information Centre. This department primarily deals with the training of customers in the use of De Beer’s products.
ADPCC: Anglo Dutch Paint Chemical Company. This is a sales organization with its own inventory, which promotes and sells the products of De Beer in Great Britain.
The managers of Human Resources, Sales & Marketing, Technology, and Projects are part of the Management Team along with the general manager (currently Operations does not have a higher-level manager). Some of the departments also have an operational manager, e.g. Production and Logistics, the Order Office and the Sales Office.
Coordination is both formal and informal. The Management Team has periodical meetings to discuss strategic and tactical issues. Operational issues are mostly dealt with by the involved departments and persons in periodical as well as ad hoc consultations.
1.3
The market
The market for car varnish can be divided in two main segments: new cars and professional car repair. De Beer aims to serve the professional car repair market, which is also called car refinish. If a company wants to supply varnish for new cars it will need approval of the car manufacturers. Besides that, the profit margins in this segment are fairly small. The car refinish segment has higher profit margins and a company does not need approval of the car manufacturers.
In the past De Beer positioned its products as the best B-brand, but this was before the company launched its water-based series. At the moment De Beer is in the middle section of A-brands.
1.3.1 Competitors
The market for car refinish has several major players. The global big four players are: Dupont, PPG Industries, Akzo Nobel (specifically its subsidiary Sikkens) and the BASF Group. Besides these four global players there are also some regional large players such as Sherwin-Williams, Nippon and Rock. Valspar and De Beer belong to a group of smaller global players along with among others Lechler, Max Meier and MIPA. Next to these large players there are also many small local companies active on the car refinish market.
1.3.2 Customers
The customers of De Beer are mainly wholesalers and distributors. They make up more than 80% of the clientele. A small part of the clientele consists of car repair companies of different sizes. The company does not sell to consumers directly. In The Netherlands De Beer uses distributors, while in the rest of Europe a wholesaler is used in each country the company is represented in. In some countries products are sold directly to car repair companies or by means of an exclusive dealer. The relationships between De Beer and its customers have a long-term nature.
1.4
The products
De Beer delivers ready for use colour mixing systems to its new customers along with approximately 40 different mixing colours. The company makes use of a computer program called Intelligent Car Refinishes Information Service (ICRIS), which is updated periodically. ICRIS is designed to provide information about De Beer products. In ICRIS colour information, product information (technical and safety information) and VOC housekeeping (Volatile Organic Compounds, also known as solvents) are stored. The program also contains an order sheet for easy ordering. All the information is stored in five different languages (English, German, French, Dutch and Spanish) so all customers are able to use the program. With the help of this program the user can make a specific colour. The program shows the amounts of different mixing colours, hardener and thinner that need to be mixed to make a colour. If the computer is connected to a weighing machine the program also indicates when the right amounts are reached during the mixing.
1.4.1 Product groups
De Beer produces many different individual products. In order to get a comprehensive overview of all these products they are divided into several main product groups. The products of De Beer can be divided in the following main product groups:
Mixing colours Hardeners Thinners Primers Surfacers Colourless varnish Miscellaneous products Mixing colours
used for trucks. The 3000-series is somewhat friendlier for the environment. The difference between these series is the different type of hardener used. The 900-series are water-based and contain less solvents.
As of January 2007, only VOC-compliant products (Volatile Organic Compounds, also known as solvents) may be sold in and imported into Member States of the European Union. Also outside the European Union sales of water-based products are planned to increase. This means that in the future the 900-series will become more important. Water-based varnish currently accounts for approximately 16% of total turnover. Furthermore, approximately 150,000 litres of water-based varnish are currently sold, which will increase to approximately 750,000 litres. This means that water-based varnish production will need to expand fivefold in the near future.
Type Series
Synthetic 200-series
Two components 400-series, 700-series and 3000-series
Base coat 500-series
Water-based 900-series
Table 1 - Subdivision of mixing colours
Hardeners and thinners
Hardeners are used for drying a varnish. Varnish needs a hardener to dry, except for the synthetic 200-series. Thinners are used to make a varnish suitable for spraying. The type of hardener and thinner depends on several factors, including temperature. In warmer countries the weather helps the varnish dry so a slower hardener is necessary than in colder countries.
Primers and surfacers
Primers are applied directly on metal in order for varnish to attach to the metal, otherwise the varnish will peel off very fast. Primers are not added if the car already has a layer of varnish. Surfacers are used to fill up and cover the rough structure of the primer. In combination with surfacers primers are a first layer.
Colourless varnish
A colourless varnish is a top layer and used for extra protection and aesthetic purposes. All products of the 500-series are covered by a colourless varnish.
Miscellaneous products
characteristics, and promotional items are part of the product range of De Beer. These are also ordered from a supplier. Some customers also order packing materials and labels from De Beer.
Figure 2 shows the layer building system, which makes clear in what way the different product groups of De Beer are used when applied to a car.
Figure 2 – Layer building system
1.4.2 Brands
RESEARCH DESIGN
This research design will explain what the research is about and why it was performed. Furthermore, it will elaborate on the way the research is shaped and what strategies were used for data gathering, data processing and data analysis. First, the initial motive of the research will be elaborated on, which explains why this research was performed. This is followed by the problem statement, consisting of the research objective, the research question, and restrictions and prerequisites of the research. Then the research method will be explained followed by the sub questions. After that, the methodology of the research will be discussed.
2.1
Initial motive
To finish the Master of Science Business Administration, specialization Operations & Supply Chains, at the University of Groningen it is required to write a final thesis. The final thesis can take the form of theoretical research, empirical research, or an analysis of a problem relating to an organization (Thesis and graduation manual for the Master’s degree programme in Business Administration). This research design is part of the latter.
The subject of the research is De Beer Lakfabrieken BV located in Lelystad. De Beer is a continuously growing car refinish manufacturer.
This research focused on improving an element of the company’s performance, namely the delivery performance. Because delivery performance was expected to be too broad a subject to deal with in the time that was allocated for this thesis, it was narrowed down to delivery reliability. In the context of this research delivery reliability is divided in three components: on time delivery, complete delivery, and delivery of the right products.
Due to a busy schedule and the relocation of the facility in Beuningen the company could not find the time to perform this research itself. By providing the possibility for a student to perform the research both parties could benefit from this research.
2.2
Problem statement
The problem statement contains the need for the research project and it has to indicate why the status quo needs to be changed (Cooper and Schindler, 2003). Its function is synchronization with the customer and internal guidance of the research (De Leeuw, 2003). The problem statement consists of a research objective, a main research question, and restrictions and prerequisites of the research.
2.2.1 Research objective
The research objective makes clear what the relevance of the research is, i.e. why the research is performed (De Leeuw, 2003). As mentioned in the previous section, De Beer is a fast growing organization with ambitious goals. In order to facilitate this growth and to reach set goals the management of De Beer is continuously searching for ways to improve performance. This research was intended to help them in this endeavour by researching the delivery reliability performance of the company and, if possible, by giving suggestions to improve it. The management of De Beer receives a document in which is described what the situation with regard to the delivery reliability performance was in 2005, which factors caused this performance, and if and how these factors can be improved.
2.2.2 Research question
The research question makes clear what question will be answered, i.e. what will be researched (De Leeuw, 2003). The following main research question gave direction to the research project:
“What are the possibilities to improve the delivery reliability performance of De Beer towards its customers?”
2.2.3 Restrictions and prerequisites
The research was subject to certain restrictions and prerequisites, set by the University of Groningen and by De Beer. These are listed in this section.
De Beer
The research will focus on the delivery performance, in particular on the delivery reliability towards customers.
The confidentiality of information will be judged by employees of De Beer. Any confidential information will be in appendices, which will not be added to the public version of the research report.
The research will start on March 7th 2006 and will be completed before August 23rd 2006, at which time the contract between both parties will end.
University of Groningen
The master thesis is part of the master’s degree programme MSc BA Operations & Supply Chains and will thus be subject to the regulations as listed in “Thesis and graduation manual for the Master’s degree programmes in Accountancy & Controlling, Business Administration, Fiscal Economics and Human Resource Management”.
The research has a credit load of 25 ECTS and will take place during a period of 22 weeks, divided in 11 weeks of 3 days and 11 weeks of 5 days.
The language used in the research report will be English. The research report will have a public nature.
2.3
Research method
The main research question aims to improve the delivery reliability performance of De Beer. However, to do this it was first necessary to determine the current performance. This presented a starting point for the research. For determining the current performance it was decided to focus on the year 2005. In this way data over an extended period of time was available, which presents a more representative view. The current performance was determined by looking at all orders received during 2005.
When the current performance was determined the next step was to find causes for the performance. This meant searching for reoccurring factors that have a negative impact on the performance. These factors were determined by means of the orders received during 2005 as well as by looking at the processes performed to fulfil the orders, i.e. the order cycle. Also, interviews with the employees and samples of activities of the order cycle further helped in finding the causes of the performance.
Figure 3 – Research method
2.4
Sub questions
In order to answer the main research question several sub questions needed to be answered. These questions address specific parts of the research and lead to an answer on the main question. These specific parts are identified and specified in the previous section. The following sub questions are derived from figure 3:
1) What does the order portfolio of 2005 look like? 2) In what way is the order cycle organized?
3) What is the current delivery reliability performance, measured over 2005? 4) What are the causes of the current delivery reliability performance?
5) In what way can these causes be dealt with in order to improve the delivery reliability performance?
2.5
Methodology
The methodology describes the research protocol in which the process of data gathering and data processing is described. It accounts for the reliability and validity of the empirical research.
2.5.1 Data gathering and processing
Groningen. The publications and books used can be found in the reference section. Next to these publications, also internal documents were used in order to research the company specific data. This company specific data was found in the database ICRIS, the ERP system BAAN IV and documents in the archive. Data about 2005 was analysed in order to determine the delivery reliability performance during that year. This gives a comprehensive picture of the performance specified to, for example, product, time period, customer region, or customer.
On the other hand field research was employed. Samples were taken of the activities of the order cycle in order to have a better understanding of the order cycle and its activities. Also, interviews played an important role during the research. Employees that deal with the processes on a daily basis know best what can go wrong in the processes and what does go wrong in the processes. They also helped to understand the different procedures and sub-processes.
All data gathered needed to be processed. Some data could directly be analysed and added to the report (e.g. interview results), while the data about orders (from BAAN, archive records and samples) were processed in an Excel-sheet. This made it possible to make calculations with the data and make graphs of it. In this way a clearer picture of the observations and the possible problems could be made.
2.5.2 Data analysis
Answering the sub questions
The data obtained from the sources mentioned in the previous section helped answering the sub questions. The data files from the ERP system made it possible to analyse the order portfolio of 2005. By transferring the data to Excel-sheets it was possible to make calculations and graphs of the data. This helped answering the first sub question.
The second sub question was answered by interviewing the employees. A general overview of the order cycle could be made by interviewing managers, after which each individual activity could be further elaborated on by also interviewing the people that perform the activity. This resulted in a description of the order cycle viewed from multiple levels.
By analysing the order portfolio in the Excel-sheets and determining criteria by means of interviews the delivery reliability performance could be determined by making the necessary calculations. This answered the third sub question.
In order to answer the fourth sub question the order cycle was used to find causes of the performance. By using the order cycle the analysis could be structured. Also, it can be assumed that causes for the performance can be found in the activities that were performed. More about the order cycle as framework can be found at the end of this section.
The order cycle as framework for the analysis
After the delivery reliability performance was analysed the causes for this performance needed to be identified. The order cycle was chosen as a primary framework for identifying these causes. Literature research revealed several factors that can influence the delivery reliability performance. All of these factors were related to (a) part(s) of the order cycle. Brown and Vastag (1992) also take a look at the entire order process, from receiving the customer order to delivery of the order, to develop several hypotheses about possible factors that have an influence on delivery reliability. Once the order cycle was described it was clear what activities are performed by De Beer to deliver an order. Looking at these activities ensures that both general factors and company-specific factors can be identified.
The order cycle is especially useful as a framework for analysing the on time delivery performance. Some activities have a standard processing time, while for other activities a maximum processing time can be determined. By analysing the normal duration of the different activities, found by combining standard and maximum processing times, and the actual duration of the activities it is possible to find out which activities take too long to perform. This indicates if the company faces problems performing these activities. This analysis can be performed for individual activities or for a cluster of activities, depending on the available data.
PART 2
ANALYSIS
THE ORDER CYCLE
The order cycle at De Beer consists of all the activities that are performed from the moment an order is received and entered into the system till the moment it is ready for transport. The transportation time is not included, because in the scope of this research the order cycle only includes internal activities since these are directly controllable by the company. Moreover, apart from standard transportation times no data is available of the actual transportation time and the moment the customer receives its order. This chapter will first discuss the different activities in the order cycle, followed by a description of the different possible order cycles.
3.1
Activities in the order cycle
3.1.1 Order acceptance and entry
If an order meets certain prerequisites it is accepted unconditionally. First of all, the customer already needs to be in the system, i.e. it needs to be a known customer. For example, if an unknown Dutch customer wants to place a small order then he will be directed to the nearest distributor. Second, the order needs to fit the usual demand pattern of the customer in question. For many known customers the system contains a list with products the customer usually orders as well as a negotiated price list, which can be established for specific customers or for a country or region. Normally, although quantities may vary, a customer orders the same type of products. If, for some reason, the customer deviates from this pattern, the sales manager responsible for the customer is consulted to find out if this deviation is a result of a special offer he made. If not, the customer is consulted to find the cause of the deviation in the demand pattern. In consultation with all parties, internally and externally, it is decided if and when the order can be delivered.
3.1.2 Planning
The Planning department is responsible for replenishing the inventory of stock items up to acceptable levels on time and scheduling production items in such a manner that it is possible to meet due dates. They receive two important signals that help them do this. First, when the inventory of a certain product diminishes below its reorder point the system will give a warning. Those products need to be replenished before the inventory is depleted. This signal can also be projected into the future, i.e. the system can indicate when the reorder point will be reached considering current demand. Second, when the new demand for a product cannot be satisfied with the current inventory or if the fulfilment of the latest demand will diminish the inventory to an unacceptable level the Order Office will give a warning that they have a shortage. Every day a MRP run is performed that is checked and, if necessary, modified by the Planning department. This run has a time fence of approximately 30 days.
Differences between product groups
Different planning situations exist for the different product groups. Of the mixing colours approximately 95% is on stock in tanks as an intermediate inventory. When there is not sufficient stock of a product in the finished goods inventory, the Planning department first checks if the product is on stock in a tank. If this is the case, it only needs to be processed by the Filling & Packing department. If the product is not on stock in a tank, the next step is to check if a paste is on stock. In that case the Mixing department processes it into a finished product, after which it can be processed by the Filling & Packing department. When nothing is on stock, the entire production process needs to be performed and all departments need to be scheduled. It is standard procedure that a new batch is scheduled for the Mixing department when a tank with a finished product is emptied by the Filling & Packing department. The same applies for the Dispersing department when a paste is used by the Mixing department. In this way the intermediate inventories for mixing colours will be kept at acceptable levels, without which it would be extremely difficult to maintain the current delivery speed and meet due dates.
No intermediate inventory exist for almost all primers and surfacers. For these products the entire production process needs to be scheduled if the finished goods inventory becomes too low. The same applies for water-based mixing colours. The hardeners, thinners and colourless varnish also do not have intermediate inventories. All these products are only made to order, replenishment orders as well as customer orders.
Planning method
100% utilization in the system does not always mean that the machine is fully loaded. This depends on other factors, e.g. number of colours and associated cleaning time. This method is also used for the Mixing department and the dissolvers in the Dispersing department. For the scheduling of the dynomills of the Dispersing department a different method is used. For each week a Microsoft Excel file is made that shows for every day which batches are scheduled. Each dynomill is used for a specific colour and only one or two batches can be produced a day on each dynomill. This is due to the long production time, which varies from 6 to 12 hours a batch.
Shop floor planning
Once the week planning is fixed the team leaders are responsible for the shop floor planning of the Filling & Packing department. They determine the shop floor planning on a weekly basis in order to be able to efficiently sequence the batches. An important goal for the team leaders is to reduce the set up times as much as possible. One aspect they take into account is the type of packing to be used. Most common are the 1 litre and the 3.5 litre cans. Normally, Mondays and Tuesdays are used to fill 3.5 litre cans, while Thursdays and Fridays are used to fill 1 litre cans. Wednesdays are used for contingencies, so if for some reason the production of 3.5 litre cans is not finished yet or if the production of 1 litre cans needs to (or can) start earlier, there will be no problem. Another aspect to take into account is the type of product. After the production of every colour the machines are cleaned, which means the sequence of colours is no issue. But in the case of metallic colours the machine does not have to be cleaned if they are sequenced from fine to rough. Therefore, all metallic colours from the week planning are clustered by the team leaders.
3.1.3 Production process
The production process of De Beer consists of three production processes and one supporting process. Dispersing, Mixing and Filling & Packing are the production processes, while QC&I is the supporting process. The production processes are performed by two production lines. The first one produces mixing colours, primers, and surfacers and performs all three sub-processes. The second one produces the so called clear products, which are colourless varnish, hardeners, and thinners. These products do not need dispersing, so this production line only contains the production processes Mixing and Filling & Packing. This is the production line that is moved from the facility in Beuningen to the facility in Lelystad. QC&I supports both production lines. The production processes are represented in figures 11 and 12.
Resource picking for production batches
different, because these are only produced to order. These orders can be a replenishment order for the finished goods inventory or a customer order. This means that there are no intermediate inventories. The inventory between the two sub-processes is emptied as soon as QC&I determines that quality standards are met.
Figure 4 – Primary process of mixing colours, primers and surfacers
Figure 5 – Primary process of clear products
Sub-process 1: Dispersing
Sub-process 2: Mixing
The next stage in the production process is Mixing. The inputs for this stage are the colour paste made by Dispersing (not for the clear products), extra thickener and additives. The company has ten mixers in the first production line, which mix the colour paste in mix colours with the desired attributes, e.g. desired viscosity and colour. The large volumes are mixed and stored in an array of approximately eighty hanging tanks, which are linked with the Filling & Packing process. The proportion of colour paste to mix colour is approximately 1:10. This means that from a 1,000 litres of colour paste approximately 10,000 litres of mix colours can be made.
Sub-process 3: Filling & packing
After the Mixing stage the product is finished and ready to be packed, which is done by the Filling & Packing stage. In this stage the product is packed in various packing materials, varying from tin cans and plastic bottles to 200 litre drums. Most commonly used are 1 litre and 3.5 litre tin cans. Before the product is poured into the packing material a filter removes the last remains of rough materials. The packing is done automatic and semi-automatic by means of several types of machines: filling machines, a filling place for drums, printers, a labelling machine, a packing machine and a palletising machine.
Sub-process 4: Quality, Control & Improvement (QC&I)
After each processing stage the QC&I department checks the quality of the product. They take a sample of the product, check its quality and determine which resources need to be added to achieve conformance to specifications. When this is done a document with the results is sent to the processing stage. If the product conforms to specification it moves to the next stage, otherwise additional processing is performed at the current stage, after which a new sample is checked by QC&I.
The customer order decoupling point (CODP)
small orders a portion will be reserved of the optimal batch size that is produced to stock, which would result in CODP 1. In the cases that the intermediate inventories are not sufficient to facilitate the production of the make-to-order products the entire production process needs to be performed starting at the resources inventory (CODP 4 in figure 4 and CODP 2 in figure 5).
3.1.4 Administration before order picking
For each day the Order Office checks which orders need to be picked that day. It is necessary to check for each order if there are any shortages, i.e. if any products are not in the finished goods inventory. This can be a case of just transferring the products from a bulk location to a pick location in the inventory, but it is also possible that the products are still in production. When everything is checked the documents for the warehouse and transfers are printed and given to the people in the warehouse.
3.1.5 Order picking
According to Schönsleben (2004, p.770), order picking is the issuance of items from stocking locations for delivery. De Beer has several stocking locations, the most important of which are outlined in table 2 (note: table 2 reflects the situation in 2005). The difference between warehouse 021, 022 and 026 needs some further explanation. Warehouse 022 is the general finished goods inventory. All produced batches of end products that are produced to stock and not for a specific customer are stored at this location. If a part of the produced batch is needed for a customer order that has already been received it still is stored at this location to await order picking. However, some batches of end products are produced to customer order. These batches are not stored in warehouse 022 but in warehouses 021 and 026 (depending on the production location). This is usually the case with larger quantities of a certain product and they are moved to warehouse 021 or 026 to await shipping.
Warehouse Description
002 a) Documentation and promotional items. b) Packing.
c) Labels.
d) Machines and parts.
021 Production items produced for a specific customer (from the Lelystad facility).
022 General finished goods inventory (not produced for a specific customer).
026 Production items produced for a specific customer (from the Beuningen facility).
The warehouse of De Beer uses a discrete picking strategy, which means that orders are picked one after the other (Schönsleben, 2004, p.770). The products are picked from the different storage locations and consolidated into one order. After this, the order awaits further administrative processing and shipment. For most European customers the picking time is set at one day. For large customers (including large European customers) and container shipments (usually outside Europe) the picking time is set at two days. These picking times are included in the standard processing time.
3.1.6 Administration after order picking
After the order is picked in the warehouse the documents are send back to the Order Office. Here they enter the real amounts that are picked and, if necessary, changes are made to the order in the system, such as quantity changes due to inventory differences or damage to products. When this is entered into the system a check is performed on any order lines that are in backorder. If these have arrived in the finished goods inventory they need to go through the order picking process again. When the order is complete or if it is decided that any shortages will not be delivered at that time, the packing slips are printed and given to the account managers. At that point the order is ready for shipment and awaits pick up by an external transportation company.
3.2
Possible order cycles
Not all activities of the order cycle need to be performed for every order. For orders that only contain products that are in the finished goods inventory planning and production activities do not need to be performed anymore. Some products need to be produced from the beginning, while others are on stock as a semi-finished product or only need to be filled and packed. When all possibilities are taken into account four order cycles can be distinguished. These order cycles are referred to as order cycle A, B, C and D (see figure 13). It is possible that multiple order cycles apply to one order, because each order line will have its own order cycle.
3.2.1 Order cycle A
Order cycle A refers to the situation when the ordered products are in the finished goods inventory at the time the order is accepted and entered into the system. This order cycle is relatively short, because after order entry only the administrative activities and the order picking need to be performed. No planning or production activities are necessary for these products.
3.2.2 Order cycle B
Order cycle B only includes the Filling & Packing department, which means that the finished product is in the intermediate inventory. This can only be the case for solvent mixing colours, because the other product groups do not have an intermediate inventory and need to be processed by the entire production process.
Figure 6 – Possible order cycles
3.2.3 Order cycle C
3.2.4 Order cycle D
THE ORDER PORTFOLIO
The order portfolio of 2005 is based on data retrieved from the ERP system and is comprised of all the orders De Beer accepted from 1 January 2005 till 31 December 2005. During this period De Beer accepted 5597 orders consisting of 91729 lines. Because this is a substantial amount of data this chapter focuses on a global analysis of the order portfolio. The remainder of this chapter is outlined as follows. First, some initial notes about the data files used are discussed. This is followed by an overview of some general characteristics of the order portfolio. After that, an overview of the most ordered product groups and products is given, followed by an elaboration on the impact of Dutch customers on De Beer. Finally, the standard processing time is discussed and analysed.
4.1
The data files
The data is retrieved from the ERP system BAAN IV. Due to the size of the data it is divided into four data files, one for each quarter of 2005. For each order line several characteristics are listed, such as dates and quantities. Not all order numbers in the data files are used in the analysis. Order numbers in the 500000 range are credit orders and are used when a customer has not received certain products that are charged. Order numbers in the 550000 range are returned cargo orders and are used when products are returned because the customer ordered different products or when products are disapproved of for quality reasons. Order numbers in the 900000 range are example orders and are used for installation orders so not everything has to be entered into the system all over again every time an installation order is received. The order lines in these ranges are linked to a ‘normal’ order line but this link cannot be extracted from the available data. Therefore, in consultation with the company, the order lines from the 500000, 550000 and 900000 range will not be part of the analysis because they would distort the results. Moreover, because they cannot be linked to a ‘normal’ order line with the available data they are not used in any analysis during this research.
in table 2. In the tables in this chapter the areas are referred to by their codes. It is also important to mention here that although customers from Holland should be assigned area code S01 almost all of them are assigned area code S03. This can be viewed as contamination of the system but it does not affect the analysis because the country of origin of all customers is known.
Table 3 – Area codes
4.2
General characteristics of the order portfolio
Figure 4 and figure 5 show respectively how many orders are received from a certain area in terms of percentage and how much is ordered from a certain area in terms of percentage.
Number of orders per area (%)
62% 3% 8% 8% 8% 1% 1% 1% 2% 3% 3% S01 S02 S03 S04 S05 S06 S07 S08 S09 S10 S11 S12
Figure 7 – Percentage of orders per area
When the two figures are compared some elements stand out. First, although area’s S07, S08, S09, S10 and S11 place relatively few orders, ranging from 1% to 3% of the total number of orders, the relative amount ordered is larger, ranging from 5% to 13% of the total amount ordered. Second, area S03 places 62% of the orders but ‘only’ orders 29% of the total amount ordered. This means that customers from area’s S07, S08, S09, S10 and S11 place orders less frequent and order, in comparison to area S03, larger amounts. This can be explained by the distance between De Beer and these customers. Products ordered are usually shipped in containers
and transported by ship. Due to the long transportation times larger amounts are ordered. In contrast, customers from area S03 place smaller orders more regularly. These orders are transported by truck over smaller distances. This is also illustrated by the weekly fixed pickup days for some European countries. Especially orders from The Netherlands, Belgium and Germany have more than one fixed pick up date.
Amount ordered per area (%)
29% 3% 13% 13% 13% 6% 5% 6% 9% 1% 2% S01 S02 S03 S04 S05 S06 S07 S08 S09 S10 S11 S12
Figure 8 – Percentage of amount ordered per area (compensated for backorders)
Another striking fact is that orders from area S02 amount to 8% of total orders received but these orders are not even 1% of the total amount ordered. The administrative activities in the order cycle for these small orders are the same as for larger orders from the other area’s. This means that orders from area S02 can have a disruptive effect on especially the administrative processes.
Amount ordered per month
0 50000 100000 150000 200000 250000 300000 350000 400000 450000 500000 550000 Janu ary Febr uary March April May June Juli Augu st Sept em ber Octo ber Nove mbe r Dec embe r
Figure 6 shows how the demand was distributed throughout the year. There were some clear peaks in the demand, especially in June. It is striking that just before the peak in June the lowest demand of the year was recorded in May. Also striking is the low demand at the beginning and the end of the year.
4.3
Most ordered product groups and products
The data files show that De Beer delivered 2137 different products. The basic products of De Beer are far less than this, but each different sized packing has its own code. For example, a 1 litre packing of a product has another product code than a 3.5 litre packing of the same product. Private labels, i.e. products with customer specific packing, also have their own code. From some products large amounts were ordered during 2005, while from other products very little was ordered. The complete list can be found in the electronic appendices for De Beer along with a list with the ordered amount per product group. To give an impression, table 3 shows the top 10 product groups and table 4 shows the top 10 products and the corresponding amount ordered.
Table 4 – Top 10 ordered product groups.
Table 5 – Top 10 ordered products.
CONFIDENTIAL
4.4
Impact of Dutch customers
De Beer fulfilled orders from 77 different countries all over the world and internal departments from the warehouse in Lelystad. Because the customers from abroad are mainly distributors it is unknown how many customers eventually use products of De Beer. However, this is not important with regard to the ordered amount of products. It is striking how important the Dutch market still is to De Beer. Some numbers are shown in table 5, which clearly show the importance of the Dutch market for De Beer. It is not surprising that fulfilling orders from The Netherlands, together with orders from Germany and Belgium, have priority over fulfilling large orders from the rest of the world. This is because customers from The Netherlands, Belgium and Germany usually order small quantities on a frequent basis. Therefore, it is possible to deliver from the finished goods inventory and assign them a standard processing time of only one day, as opposed to a standard processing time varying from 5 to 15 workdays for most other customers.
Total Holland % of total
Customers 234 53 22,65%
Orders received 5597 2080 37,16%
Quantity ordered 4225192 464557 10,99%
Table 6 – Impact of Dutch customers.
4.5
Standard processing time
Every customer is assigned a standard processing time. This is determined based on the size of an average order and the probability of production items. If an order is smaller than normal and/or does not contain production items the processing time will be shorter. Included in the standard processing time is the production time, the picking time and the administrative processing time. It is used to determine the planned order ready date at the moment of order entry. The order ready date is further explained in section 5.2.1. De Beer uses four standard processing times: one, five, ten and fifteen workdays. As stated above, these standard processing times will not always apply, depending on the actual order received. Figures 7 to 10 show how often the different standard processing times actually are met. They show a distribution of how many days the Order Office thinks to need to deliver an order. This is done by checking how many days are between the order date and the planned order ready date for each order line.
Figure 10 – Planned processing time for all order lines with a standard processing time of one day
Figure 8 shows the same as the previous figure only then for the order lines with a standard processing time of five days. A reasonable number of order lines is indeed delivered within the five days. However, a large part takes slightly longer. Most order lines in this category need between five and eleven days to process, probably caused by the regular presence of production items in the orders.
Figures 9 and 10 show the situation for order lines with a standard processing time of ten days and fifteen days respectively. It is interesting to see that in these categories a large number of order lines have a large expected processing time. In contrast to the previous two figures these graphs show a far more spread out distribution. Customers in these categories probably place larger orders with many production items. The figures show that in these categories the standard processing times are often not met.
Figure 12 – Planned processing time for all order lines with a standard processing time of ten days
DELIVERY RELIABILITY PERFORMANCE
In this chapter the performance of De Beer regarding delivery reliability towards its customers during 2005 will be analysed. First, the concept of delivery reliability as well as ways to measure it will be explained. This includes a description of how the concept will be used in the context of this research. After that, the components of delivery reliability distinguished in this research will be elaborated on. These components are on time delivery, complete delivery and right delivery.
5.1
Delivery reliability and ways to measure it
Theory on manufacturing strategy advocates the development of manufacturing capabilities in order to support or provide a competitive advantage in the marketplace. Literature on the subject states the need for the manufacturing function to develop bases of competition linked to a corresponding marketing and business strategy. One of the bases of competition identified is delivery (Handfield and Pannesi, 1992). Although many authors refer to delivery, others distinguish between delivery speed and delivery reliability/dependability. Slack and Lewis (2002), for example, identify five generic dimensions of a company’s performance with which it will attempt to satisfy market requirements. They call these dimensions performance objectives and their purpose is to articulate market requirements in a way that will be useful to operations. Slack et al. (2004) describe these five generic performance objectives as follows:
Quality: doing things right. Speed: doing things fast.
Dependability: doing things on time. Flexibility: being able to change. Cost: doing things cheaply.
concept is expanded to include three components: on time delivery, complete delivery, and delivery of the right products. In this way explicit attention is given to the components of complete and right delivery, instead of assuming that these objectives are met when measuring dependability.
Delivery reliability is an important factor for manufacturers and, according to Berry et al. (1990) and Roth et al. (1992), has been demonstrated to influence customer satisfaction, as well as return on investment. A number of surveys of buyers' attitudes, e.g. Cunningham and White (1973) and Cunningham and Roberts (1974), have concluded that delivery speed and reliability are essential features of a supplier's offering for which many buyers are prepared to pay a higher price (Cunningham and Dale, 1983).
According to Graves (1981), a company’s ability to deliver a product reliably can be represented by a number of measures, including, number of on time deliveries, percent of orders that have later than promised delivery dates, average lateness overall, or average lateness of late orders (Brown and Vastag, 1992).
5.2
On time delivery
On time delivery means delivering the ordered products on or before the promised delivery date. Although for some customers delivery before the promised delivery date can be unacceptable, this will not be considered in this research. The main reason for this is that if the company is able to deliver early, it is assumed that it is also able to deliver at the exact promised delivery date. Furthermore, it is not known for all customers if early delivery is acceptable or not. For this research on time delivery is measured by means of the number of on time order lines and the percentage of order lines that have later than promised delivery dates. In addition, the tardiness of late order lines is also measured, i.e. how long after the due date are late order lines delivered. This is measured by counting the workdays between the planned order ready date and the realized order ready date (realized picking date + an extra administration day: this is further explained in section 5.2.1). A reason for focusing on order lines and not on whole orders is that there can be more than one promised delivery date for an order. It is possible that some order lines of an order have a different promised delivery date than the other lines of the order. Furthermore, focusing on order lines gives the opportunity to retrieve information that will be harder to obtain on a higher aggregation level, e.g. with regard to product groups and products.
5.2.1 Criterion for being on time or late
control of the company. Therefore, it is necessary to find a different way to measure on time delivery, i.e. an internal due date that is known for all orders. For this purpose figure 14 shows a timeline from the moment the order is entered into the system after acceptance till the customer receives the products. The distances between the dates are arbitrary and do not reflect the real amount of time between the dates. The timeline only reflects the chronological relationship between the dates. To avoid misunderstandings the timeline will receive some further explanation.
Figure 14 – Timeline from the moment an order is placed till it is received by the customer
When an order is received, it is almost always accepted and entered into the system on the same day. It is possible that in some cases an order is entered into the system on the next day, but these cases are exceptions. The order date is the date the order is entered into the system, which is recorded for every order along with the planned picking date and the planned order ready date. This is the starting date for the order processing time. When the products are in the finished goods inventory or when they have been produced, they are ready to be collected from the storage locations. This happens on the picking date. Although the timeline only shows the picking date, the system contains two dates: the planned start date for order picking and the date the actual order picking is finished. After the order picking and the necessary administration activities are completed the order is ready for shipment to the customer. This date is referred to as the order ready date. The system only contains the planned date on which the order has to be ready, which is determined at the moment of order entry based on the standard processing time for the customer in question. After the order is ready for shipment it is picked up for transport by an external transportation company and eventually arrives at the customer. Between the order ready date and the pick up date the invoice is sent to the customer.
shipment, then the order is regarded as being on time. For every other customer the administration activities after order picking are completed on the following day. In these cases the realised picking date needs to be at least one day before the planned date on which the order needs to be ready for shipment to be regarded as being on time. Summarized this means:
It is important to note that backorders are not mentioned specifically with regard to late delivery. According to Schönsleben (2004, p.240), a backorder is an unfilled customer order or commitment, an immediate (or past due) demand against an item whose inventory is insufficient to satisfy demand. So a backorder does not necessarily mean that the order is not delivered on time. This corresponds with the use of backorders at De Beer. When not all products are ready on the planned picking date, e.g. due to a delay in production, the products that are not ready are classified as a backorder. It is possible, however, that the products are ready before the order ready date and are still on time. Therefore, the criterion mentioned in this section for on time delivery is comprehensive because it includes the backorders that are ready on time and does not include backorders that are not ready on time.
5.2.2 Late order lines
When the criterion for being on time or late is applied to the order lines in the four data files an overview is created of the number of order lines that is on time and the number of order lines that is late for each quarter of 2005. This overview is shown in table 7. The overview shows that 72686 of the 91729 order lines were delivered on time, which corresponds to 79.24% of the total number of order lines received. Furthermore, 19038 of the 91729 order lines were not ready for shipment on time using the criterion mentioned earlier, which corresponds to 20.75% of the total number of order lines received. It is these order lines that are of interest in this research and need further analysis. Q1 Q2 Q3 Q4 Total Total 21291 25025 23845 21568 91729 On time 16683 18027 19833 18143 72686 Late 4607 6996 4010 3425 19038 Unknown 1 2 2 0 5
Table 7 – Number of order lines divided in on time and late - Customers from The Netherlands, Belgium and Germany -
Realised picking date ≤ planned order ready date => on time delivery
- Customers from the rest of the world -
A more detailed overview in table 8 shows from which areas the 19038 late order lines were received. In addition, for each area it is shown how many days the order lines were late. A large part of the late order lines are from areas S03 and S05 (the geographical regions that correspond to the used area codes can be found in table 2). Areas S02 and S06 also show a considerable amount of late order lines. Together these four areas account for almost 70% of all late order lines. A striking fact is that 51% of all late order lines is only one day late. Also, 9.2% of all late order lines is more than a work week late, which accounts for 1.9% of all order lines received.
On the one hand, depending on priorities, the late order lines from area S02 can be less important to improve for De Beer because these are internal orders, which are usually very small and have almost no direct impact on the perception of external customers. On the other hand, internal orders are also important because the products are used by other departments, e.g. for improving quality or performance and for customer relations.
Area 1 day late 2 days late 3 days late 4 days late 5 days late 6-10 days late 11-15 days late >15 days late Total late 2005 % of total late 2005 S01 0 0 0 0 0 1 0 0 1 0.01% S02 1325 309 89 98 269 92 48 18 2248 11.81% S03 2088 585 237 288 317 436 188 176 4315 22.67% S04 156 181 82 12 1 8 10 16 466 2.45% S05 1900 1239 773 93 138 126 63 114 4446 23.35% S06 1736 380 28 16 19 28 36 34 2277 11.96% S07 953 270 226 16 31 9 3 41 1549 8.14% S08 428 180 136 65 55 4 4 0 872 4.58% S09 694 173 175 205 80 141 4 0 1472 7.73% S10 123 200 127 88 118 23 5 0 684 3.59% S11 166 158 54 6 1 104 3 19 511 2.68% S12 146 31 10 3 5 2 0 0 197 1.03% Total 9715 3706 1937 890 1034 974 364 418 19038 100%
Table 8 – Number of late order lines per area divided to number of days late
Next to dividing the late order lines to area, it is also possible to divide them according to product group. This is shown in table 9. Again, for each product group it is shown how many days the order lines were late. The product groups that stand out are 103000, 273000 and 275000. Together these three product groups account for more than 53% of the total late order lines. This is a large percentage considering the company uses more than 50 product groups.
