Creating transparency
A case study at Tomin Metaal to realize transparency of the primary process
University of Groningen
Faculty Economics and Business Master of Technology Management
Author: Anne-marie van Dijk
Student number: 1137808
Date: 10-09-08
Supervisors Tomingroep Dhr. J. Postma Dhr. R.A. Wiersma
Supervisor RuG: Dr. J.A.C. Bokhorst
“Van IJzer wordt je niet Wijzer,
alleen maar Grijzer” (Jan Bijlard)
“Als het goed is wijst het gebouw je de weg” (Victor Klein)
“…. anders gaan we de Bietenbrug op” (Ronald Wiersma)
“Klanten moet je aaien,
aandacht, affectie en interesse” (Pieter Atema)
“Ik heb een probleem… “Spoed is slecht gepland”
Nee, je hebt een uitdaging”
(Productiebureau Metaal)
“Als je doet wat je deed,
zul je krijgen wat je hebt” (Pieter Atema)
Ik heb geen collega’s,
ik heb alleen maar concurrenten
(Jan Heeres)
“Je kunt je kennis verdubbelen door het te delen” (Arjan Ploeg)
“Een trein moet je laten lopen”
(Koen Hoetmer)
Preface
With this thesis I present the results of a five months during research at Tomin Metaal in Hilversum, part of the Tomingroep. The research is meant complete the master Technology Management. At a lean workshop of the Lean Operation Research Centre (LORC) I contacted Jelle Postma the business unit manager of Tomingroep Industrie. He offered me the opportunity to perform my research at a versatile environment. Besides the subject of my thesis I learned a lot about different processes in the company. I experienced the period as interesting, exciting and highly informative.
I would like to thank Tomin Metaal for providing me with this opportunity, the freedom and the responsibility necessary to complete my thesis. During my research at Tomin Metaal my ‘colleagues’ always were prepared to discuss ideas and information for my thesis. I am also grateful for the time and patience the operators had with me and my questions. Especially I would like to thank Jelle Postma and Ronald Wiersma, my supervisors at Tomin Metaal, for dedicating time for reviewing my thesis and supporting me with advice. Furthermore I would like to thank my thesis supervisor, Jos Bokhorst, for his constructive feedback, useful tips and endless patience. Finally, I would like to thank my parents and boyfriend for their continuous support.
Groningen, September 2008
Executive Summary
Transparency is needed to control a process, this can be realized by understanding what happens in the process. Tomin Metaal, part of the Tomingroep, is a company which sales have increased during the last years and they have become too large to keep the process transparent. This thesis explores which parts of the process can be structured and how these can be controlled by using the following research question:
“ Which factors influence the transparency of the primary process of Tomin Metaal and which improvements can be made to increase transparency, so it becomes a controllable process? ”
To answer this question this thesis focuses on the planning, the warehouse and the production process of a Bovema flap. Empirical research and theoretical research are used to identify the current performance and a new structured proposal. The possible solutions and recommendations are based on lean principles. The fact that Tomingroep is a sheltered workshop is if necessary taken into account with the recommendations.
The lead time and the amount of work in process (WIP) are the repeating variables which tell a lot about the transparency of the process. The current planning does not give a clear insight in the amount of WIP and also not to the needed or available capacity, besides it costs a lot of effort to determine the length of the lead-time and the continuity of an order. This makes it hard to control and not transparent.
In recommending a capacity plan should be introduced to create insight in over- and undercapacity. This leads to transparency and Tomin Metaal will be able to rearrange the planning, the occupation of the machines or the position of the operators. In a next stadium it can be advised to implement a more extensive planning system like the lean planning and control system (LPCS) introduced by Slomp et al (2008). This system leads to a constant, and as small as possible, amount of WIP.
At the moment the warehouse has a hard time to control their processes. This is caused by the lack of registration and the many tasks the warehouse employees have to perform. By splitting the internal logistics from the control of in- and outgoing materials, the control and the transparency will be improved. In the pilot the spot-welding process is changed so it became possible to realize a lead time which equals the bare spot welding process. Due to the pilot it seems possible to increase the output of spot-welded flaps with 67%. The involvement of the operators and their supervisors have a major influence on the continuation of the pilot.
A low amount of WIP has a positive influence at the lead time of a process. But the amount WIP is often used to cover problems (river of inventory). By solving the problems or optimizing the production process it will be able to lower the amount of WIP even more. An action research has been done of the Bovema case to analyze which possibilities Tomin Metaal has to improve a process and to give advise about how to implement a new idea or process.
Index
1. INTRODUCTION ... 7
1.1 Outline... 7
2. TOMINGROEP METAL DIVISION... 8
2.1 Tomingroep ... 8 2.2 Department Metaal... 8 3. RESEARCH IDENTIFICATION ... 9 3.1 Introduction... 9 3.2 Black-box ... 9 3.3 Warehouse space... 9 3.4 INK-management model ... 10 3.4.1 Continuous improvement ... 10 3.4.2 Transparency ... 10 3.4.3 Result orientated ... 11 4. RESEARCH FRAMEWORK ... 12 4.1 Objective ... 12 4.2 Problem definition... 12 4.3 Conceptual model... 13 4.4 Sub questions ... 14 4.5 Research design... 14 5. PLANNING... 16
5.1 Description Current Planning Process... 16
5.1.1 Flow of information necessary for the planning ... 18
5.1.2 Flow of information necessary to know the position of work in process... 20
5.2 Performance Current Planning Process... 21
5.2.1 Lead time... 22
5.2.2 Work in Process ... 25
5.3 Theoretical framework ... 27
5.4 Conclusion of the diagnosis ... 28
5.4.1 Conclusion of the current communication performance ... 29
5.5 Design of a capacity planning ... 29
5.6 Available planning theory ... 31
5.6.1 Workload control (WLC) ... 31
5.6.2 Kanban... 31
5.6.3 POLCA... 32
5.6.4 CONWIP ... 32
5.6.5 Lean planning and control system (LPCS) ... 32
5.6.6 Results planning... 33 6. WAREHOUSE... 34 6.1 Responsibilities warehouse ... 34 6.2 Performance ... 35 6.2.1 Storage space ... 36 6.2.2 Work in process... 36
6.3 Possibilities for improvement ... 37
6.3.1 Stock of materials... 37
6.3.2 Internal logistics ... 38
6.3.3 Storage space ... 38
6.4 Redesign Warehouse Arrangement... 39
6.4.1 Stock of material ... 40
7. PRODUCTION PROCESS BOVEMA FLAP ... 42
7.1 Bovema order ... 42
7.2 Process description... 42
7.3 Bovema pilot ... 45
7.3.1 Changing the process... 45
7.3.2 Pilot: changing the spot-welding process. ... 46
7.3.3 Results ... 48
8 CONCLUSION & RECOMMENDATIONS ... 52
8.1 Conclusion ... 52
8.2 Action plan... 55
REFERENCES ... 56
APPENDIX 1 ORGANIZATION CHART ... 58
A1.1Tomingroep ... 58
A1.2 Tomin Metaal ... 59
APPENDIX 2 INK-MANAGEMENT MODEL... 60
APPENDIX 3 OPERATION LIST ... 62
APPENDIX 4 NTW CALCULATIONS BOVEMA ... 63
1. Introduction
Every company wants to perform as good as possible. Strategies of companies are based on their life stage, growth and objectives. Tomin Metaal, part of the Tomingroep, is a company which sales have increased during the last years. Tomin Metaal is still working with the original production arrangements, but they have become too large to keep the process transparent. For controlling the processes, transparency is needed and it is necessary to understand what happens in these processes to control the processes. Other often heard complaints are the lack of space to store materials. These materials vary from stock materials to finished orders. These two barriers prevent further development and are the primary reasons for this research. Besides removing these barriers, the management of Tomin wants to introduce lean principles at Tomin Metaal. They expect that this will optimize the processes and improve sales.
1.1 Outline
2. Tomingroep metal division
2.1 Tomingroep
Tomin Metaal, the metal division, is one of the ten business units of the Tomingroep. Tomingroep offers industrial as well as personal services (organization chart, appendix 1). These activities occur in the domain of retail and building construction, both in conformity with the local market. Employees of the Tomingroep preferably are individuals with a distance to the labor market. Tomingroep aims to have at least a balanced trading result.
Because Tomingroep works with people who have a hard time to find a job in a regular company, it is called a SW-company, also known as a sheltered workshop. The major part of the employees has a contract based on the law of sheltered employment (Wsw). Besides this group of employees, there is also a group that participates in a reintegration project. The remaining employees are BV’ers, this group of employees have a so called regular job. They mainly fulfill all supporting functions within Tomingroep at departments such as Personnel & Organization, Salary administration and the production offices. At Tomin Metaal 85% of the workforce consists of WSW employees. There are 115 individuals in operation, 99 of them having a Wsw-social security.
2.2 Department Metaal
Tomin Metaal concerns the machining process and assembly of metal parts. For several customers they realize various machining processes for stainless steel, metal and aluminum. The currently delivered activities at Tomin Metal are:
Metal machining process
- Pretreatment (sawing, cutting)
- Plate machining process (punching, setting)
- Drilling (turn-, milling cutter- and drill work) - Welding (MIG-, TIG-, spot welding & solder) Assembly
- Mechanical assembly
- Electro assembly
- Remaining assembly activities
- Packaging /sending
Reconditioning
- Revise of several (support)equipments
- Modification of parts
Tomin Metaal is situated in Hilversum and Almere. The major part of the activities are performed in Hilversum. Almere is a smaller operating location and functions almost independently. They specialize in revising of rehabilitation resources and battery-chargers and the charging of carbon filters.
3. Research identification
3.1 Introduction
At the start of this research the main objective is:
“Create more transparency in the production process to improve the efficiency of Tomin Metaal, with a focus at the warehouse and logistics”
The research will be conducted at the metal machinery division (nr. 170) of Hilversum (appendix A1.2). This chapter will provide a detailed description of reasons for this research proposal.
3.2 Black-box
In this thesis a black-box is the description of a process of which the input and output is known, but that it is not (easily) to trace how the output is realized. The main disadvantage of a black-box is, that if you do not know how the output is realized, it is often also hard to influence and control the output. The elimination of black-box processes, so discover what happens within the processes, is a possibility to create transparency.
The production process of Tomin Metaal is currently a black-box. The management of Tomin Metaal knows what should be produced and which materials are bought and will be delivered, but what is done in between and when this is performed is mostly unknown. Information is not always reliable because it sometimes takes a while before the delivery confirmation is registered. Besides the actual amount of material does not correspond with the administrated stock. It concerns purchased sheet metal and staff material, but also materials for production delivered by customers. This difference has several causes:
- Insufficient knowledge of the products; it is not always noticed if the delivery is wrong or if the amount is not correct.
- The stored materials are mixed which results sometimes in the use of a wrong type of material.
- It is not always registered when a new sheet of metal is used at the moment the sheet being processed is damaged.
And there probably could be mentioned more reasons. 3.3 Warehouse space
The warehouse is also a part of the primary process of Tomin Metaal. During the time of this research Tomin Metaal reorganized the arrangement of the shelves in the warehouse in order to increase the efficiency of the warehouse, in this situation this means decreasing the time needed for each job. Since March 2008 Tomin Metaal (Hilversum) also has the availability of a Romney shed in which a part of the stock can be stored, this will mainly be used to store finished products.
Although the organization tries to solve the problem partly by creating ‘new’ space, the warehouse and logistics will have a focus in this research. At Tomin Metaal, the business manager and different employees do have the perception that the production process can be more efficient and that a lot of mistakes can be eliminated.
3.4 INK-management model
The Business unit manager (Bum) uses the so-called INK-management model to position and improve the organization. In previous research at Tomin Verpakken this model was also used and produced good results. Therefore the management is interested in the use of this model for Tomin Metaal too.
The INK-management model consists of several phases. The final goal is to reach the phase in which the company excels in operation. The INK-management model triggers the Bum to improve the business part, for example by improving efficiency. At the moment the insufficient measurability of the results and the processes prevents the development of Tomin Metaal. The INK is a foundation with the goal to improve the quality of the management based on the INK-management model.1 The INK-management model can be found in appendix 2.
There are three main areas which need improvements to make it possible for Tomin Metaal to move through to the next stage: continuous improvement, transparency and result oriented. Tomin Metaal is mainly situated in the first stage; activity oriented. In these situations each individual has to make sure that the work will be finished as good as possible. Expertise is highly valued and is supported by education. The organization tries to solve possible complains. The first goal is to reach stage two; process oriented.
3.4.1 Continuous improvement
During the primary process of Tomin Metaal mistakes often are solved ad hoc. Due to this type of action it is neglected to register these problems with their associated solutions. In the first phase of continuous improvement, improvements and actions for recovery are only made after a mistake has been identified. This can be compared with the kind of performance at Tomin Metaal, but can be improved by introducing project-teams who will focus merely at the cause of mistake, so these mistakes will not happen again in the future.
3.4.2 Transparency
With a lot of processes Tomin Metaal is not aware of the performance, this means that the lead time is determined and not calculated. Also the amount of work in process is not known which has the consequence that calculation can only be based at the output and the processes are hard to influence. These processes can be seen as a black-box and because at this moment it is not visible which variables cause the results, are hard to change This is a characteristic property of the lack of transparency in the first stage. Based on interviews and information of the quality manual, some differences can be identified, these differences are part of the current performance and described in several chapters of this report. Because Tomin Metaal would like to influence their processes in order to realize a better performance, the creation of transparency will be the main focus of this report. In theory it can also be concluded that they perform well
1
at the moment, but Tomin Metaal is not able to deduct these data. So at least the communication should change.
3.4.3 Result orientated
If Tomin Metaal wants to flow to the next stage of the INK-model, they should accomplish performance measurements and take action if deviations of the performance in the primary process are recognized. Also demands of external customers should be fulfilled. At the moment Tomin Metaal uses calculations to get an impression of the results; this information is used in returning orders. Besides it should happen that if an operator or their supervisor recognizes deviations in the prescribed and the real status of work in process they should take action. These actions can vary from redoing the order; producing a larger batch size of the order, because a few parts where rejected or just inform the production office.
Of the three described variables of the INK model, the need to create transparency has priority. During this thesis this will be the first objective, also because the other two variables take advantage of a transparent process. Within continuous improvement causes of problems and mistakes can be identified and the variance in results should be reduced, which is easier to accomplish if the process is transparent, so it will become known how the results are established. This will increase the performance and a result oriented company should benefit from both (figure 3.1). Due to transparency there is known how the results are established, a workgroup continuous improvement is improving the results and the results oriented view of the company should notice if there are deviations and aim for new (continuous) improvements. Only in a few paragraphs and in the conclusion there is described how the realized and proposed changes influence continuous improvement and result orientated processes. And what Tomin Metaal should do to reach at least the process oriented phase.
4. Research framework
4.1 Objective
The aim is to create more transparency in the primary process, with a focus on work in process. So it becomes clear who has which responsibility, work efficiency will improve and no actions be performed twice. The different bottlenecks must be identified and quantified, so that it is possible for management and operators to control work in process.
4.2 Problem definition
The problem definition phrases the main question in research accessible terms which suits with the objective. It should be the most important lead for the formulation of the sub-questions. (De Leeuw, 2001).
“ Which factors influence the transparency of the primary process of Tomin Metaal and which improvements can be made to increase transparency, so it becomes a controllable process? ”
The primary process starts at the warehouse where the orders are released with the use of a pick-list. The process ends when an order is finished and the finished product is packed, this also happens at the warehouse. During this research, the focus lies at the work in process (WIP), this becomes feasible by the clogged corridors. One of the causes of these clogged corridors is that an order in production sometimes is interrupted and stored away. The space to store the materials at the workplace is not sufficient. There are not enough permanent locations to store work in process, so products often have to be located again to re-assign them into the production process and the transportation time is double. In a perfect situation the production orders will not have to be interrupted, because the order will be produced in a flow. So in theory these orders do not need space to store the WIP. Of course it is hard to reach this stage, but if the continuity of the flow could increase, the amount of WIP orders stored at the corridors should decrease.
A process has to be mastered before it is possible to control. According to the dictionary ‘Van Dale’ to be master of the process is ‘having knowledge about’2, this could only be realized for the complete organization if the process is transparent, but it also has to become measurable. Because a process can only be controlled if it is understood.
2
4.3 Conceptual model
Figure 4.1 Conceptual Model
Central in the conceptual model is the amount of work in process. In this report the amount of work in process starts if the order is released at the production floor by the production planning and the first operation is able to start.
The conceptual model presents three variables which influence the amount of work in process. The more transparent each variable is the more positive the influence at the amount of work in process. The way of planning has a direct influence at the amount of work in process. This is caused by the order release, so the number of orders in production and the continuity of these orders during production.
The internal logistics in the warehouse does also influence the amount of work in process. If this is not organized well enough it sometimes can take a while before it is noticed that an order can be started or proceeded. Another possibility why the warehouse influences the amount of work in process is that if it is not registered some orders use extra materials, another or the same order will in the end be short in material. The consequence of this can be that an order is stopped and a new order is started, which leads to extra work in process.
4.4 Sub questions
The sub questions focus on the three main variables of the conceptual model. At first the planning will be analyzed, because the way of order release and the capacity planning are the main cause of the amount of Work in Process (WIP). The second research question describes the warehouse, both internal and external logistics. If this performance suits perfectly to the planning, less waste will be created and the lead time of an operation is not disturbed by the performance of the warehouse arrangements. At last the performance of a production process will be analyzed. In this analysis the performance was measured based on output, but also throughput time of an operation. This sub question is not reducing the amount of work in process, but it optimizes the process, why it will be easier to work with a lower amount of work in process. The conceptual model shows the focus of this research. Based on this model and choices made earlier in this paper, the following sub questions can be described:
1. “Which aspects of the planning system at Tomin Metaal influence the controllability of the amount of work in process?”
a. How is the communication around the input and output of the production planning, can improvements be made?
b. How can the lead time be calculated based on the planning data and what is the lead time performance?
c. What is the average amount of work in process and what can be concluded from this performance?
2. “Which role has the stock of materials and activities of the internal logistics on the work in process?”
a. How often is the space to store materials insufficient?
3. “What is the current performance of the Bovema production process and is the performance sufficient enough?”
a. What is the current average production lead time of the Bovema production process and how can this be improved?
b. Can the performance of the Bovema order be improved? 4.5 Research design
Before the research question could be answered, the current primary process has to be analyzed. The primary process is performed at the warehouse and the production floor. The primary process is controlled by the production office that is in charge of the information of an order and the chief of production who is in charge of the planning of the orders. In practice this beholds the overview of when an order has to be finished and at which operations the orders are stationed. The primary process contains the following parts:
- Warehouse; one chief and three operators are responsible for inventory. They have to store and administrate incoming materials and outgoing materials, these materials can also be finished or semi-finished products. Besides the internal and external logistics, they have to wrap finished products in order to make them ready for transport.
department. At the moment an order is finished, the warehouse will be called in again. The welding and the drilling department both have a supervisor with both 6 operators. The department chief fulfills this job at the mechanical engineering, together with the sawmill he has 9 operators to supervise.
Several orders have to be outsourced before the product is produced according to customer demand, this is part of the primary process, but is not taken into account in this thesis, because this is hard to influence. Important parts of the primary process are the order release and the capacity planning, because this has a large influence at the work in process. A map of the location of the primary process, the work floor, can be found in appendix 1.3
5. Planning
In this chapter the way of planning at Tomin Metaal and the efficiency of order release and their effects will be discussed. Besides it should answer the question:
“Which aspects of the planning system at
Tomin Metaal influence the
controllability of the amount of work in process?”
I started with these elements because, as shown in the conceptual model, these elements directly influence the amount of work in process, which again has influence at the other subjects of the conceptual model. Interviews were used to understand the current planning process, who is responsible, which role has the planning and how and by whom can this be influenced. These results are compared with literature to be able to introduce well funded recommendations. To obtain an indication of the current performance the planning and changes in the planning were observed and with the use of calculations conclusions are formed. Based on the planning it is hard to determine the number of orders waiting at the work floor, a work sample has been made to get an indication of the number of orders waiting at the mechanical engineering and the welding department.
To realize transparency, it is important that the stakeholders are aware of their tasks, responsibilities and their powers. Communication is also supportive to the planning and its aspects, because if the communication is clear and unambiguous, fewer mistakes will be made. To give a good advice in the conclusion the current communication flow must be mapped and it must become possible to answer the following sub question:
How is the communication around the input and output of the production planning, can improvements be made?
Through the use of interviews, observations and data from the manuals, the diverse flows of information can be visualized. In this chapter also the explanation of the current information flows about planning and possibilities for improvement will be discussed.
There are a few similarities in the different information flows. In all situations the arrows present the direction of the information flow. The kind of communication can vary between face to face, phone, e-mail and internal post
The stakeholders in the flow diagrams can be compared with several stakeholders identified at the organization chart (Appendix 1).
5.1 Description Current Planning Process
position of work in process, by observing which operations are already completed and what still has to be done. This information is not automatically registered in the planning, but the planner tries to keep this up at daily basis. This system works very well with small batches, but there is no prescription for orders that are in process at different operations simultaneously. The shop floor of Tomin Metaal is expanding, this asks for a more formalized approach to handle the increased complexity (Henrich et al., 2004).
The planning performance is analyzed by observing the planning used at Tomin. This is an Excel planning which includes all operations of the metal machinery inclusive possible outsourcing addresses, see figure 5.1. The 1 in this planning means that an operation is completed. Because all operations are shown in the planning this type of planning could be seen as a detail planning.
In the current situation there are four planning criteria which are leading for the ranking of the orders at the different work-stations.
1. Workload of each employee, does every employee have a job to do? 2. Utilization of the machines
3. Delivery time.
4. Importance customer.
Based on these criteria it can be concluded that the planner is focused mainly at the degree of occupation. The planner does not want operators become idle or that the main production machines are not occupied. These first two criteria often correspond. Sales, at this moment the production office, want the planner to focus at the delivery time. If not all due dates can be accomplished, the importance of the customer is decisive. This last standard is used at the moment that planning targets seem unachievable.
During the weekly planning-meetings between the planner and the chief of the production office only the orders with deviations are discussed. This could vary from why an order is not started yet (although it should be) to the material of an order arriving too late. In between the planner gives a signal about which order should have priority when he thinks that not all deadlines could be realized. This often happens about two weeks before the deadline.
PLAN- DAT. OMSCHRIJVING METAAL BEW. A A N T A L LEVER- TIJD M a g a zi jn V o o rb ew er k in g P la a tw er k er ij A m a d a B o o rd er ij L a ss er ij U it b es te d er M o n ta g e G er ee d 26-02-08 179703 Statief T Haspelk GEREED 5 week 23 1 1 . . 1 1 1 1 27-05-08 Statief H haspelkast 790x790 10 week 28 0 0 . . 0 0 0 0 21-05-08 032904 Lasbalkstrip 620-MG BOREN 500 week 24 1 . 1 . 0 . 0 03-06-08 001801 Modificeren oph plaat 608 week 24 1 . 0 . . . 0 11-04-08 Gev. Voorstang 290422 GEREED 700 week 23 1 . 1 . . . 1 11-04-08 Gev. Achterstang 290401 ONTVETTEN 100 week 23 1 . 1 . . . 0 11-04-08 Gev. Achterstang 290402 ONTVETTEN 500 week 23 1 . 1 . . . 0 11-04-08 Gev. Achterstang 299401 ONTVETTEN 200 week 23 1 . 1 . . . 0 03-03-08 Alu-1 klep TMS-D Type 25 PLAAT/LAS 2500 week 14 1 1 0 0 . 0 0 0 25-04-08 Alu-1 klep TMS-D Type 25 KNIPPEN 990 week 23 1 0 0 0 . 0 0 0 23-04-08 Alu-1 klep TMS-D Type 25 690 week 28 0 0 0 0 . 0 0 0 17-04-08 Alu-2 klep TMS-D type 25 PLAATBEW 1814 week 20 1 1 0 0 . . 0 21-05-08 10013 RVS Bank 60 week 29 0 0 0 0 . 0 0 08-05-09 Spring ring TX 1402-60-0100 GEREED 36 week 21 1 1 . . . 1 1 08-05-09 Spring ring TX 1402-72-120 GEREED 30 week 21 1 1 . . . 1 1
Figure 5.1 Example planning Tomin Metaal. . = in production, . = exceeded delivery time . = must be finished this week.
5.1.1 Flow of information necessary for the planning
In figure 5.2 can be seen that although several participants have access, only the production office actually uses the information from the planning. For important and precise information the production office always checks or confirms the excel planning with the planner or department chief.
Based on divers input from production and the production office combined with his own observation and knowledge the planner is taking care of the input of the planning Excel sheet. For the regular orders the communication flow follows the hierarchic path. The planner and the production office inform the department chief about the actions which should be made according to the planning, he informs the supervisor and so on. As can be seen in the right lower corner of figure 5.2 in some situations the decision at which order an operator has to work, is influenced by many different stakeholders. This is due to urgent jobs, which are discussed by the planner and the chief of the production office at daily basis. But in practice the production office, the planner and the department chief have different privileges which they ventilate to the supervisors and operators.
The elaborate communication flow used for information about the planning is due to differences in perceived importance of the orders. This is caused by lack of discussion or consensus between the planner, the production office and the department chief. In this structure they do not perform according to clear responsibilities. In the current situation it is not easy to determine the status of an order, so the location of WIP, which also leads to a lot of communication. This is partly due to the location of the operation lists in large batch orders. There is one list available which can only be located at one pallet, so at one operation. The other pallets must be marked and the department chief or the chief of production must remember which operations already started and if not documented it is possible to lose data. This is why the acquired data is often checked by the production office.
5.1.2 Flow of information necessary to know the position of work in process
The information flow concerning the work in process shows similarities with the information flow of the planning. This merely concerns the communication between the planner and the production office either with or without the use of the planning system. The production office uses this information to give feedback to their customers. As can be seen in figure 5.3, there are three ‘self informing’ arrows. The first two, “a”, check if the perceived information is correct. Arrow “b” describes the communication between the operators of the different departments; they inform not only their supervisor, but also each other about the stadium of the work in process.
The warehouse and the planner have a central position in the communication flow. The planner has this position because he has to update the planning system. Because of his role as chief of production he should have enough information about the work in process. The warehouse is an important actor in the information flow, because this is where the work in process starts and ends. There are a lot of arrows directed to the warehouse. Most of this communication is face to face or by phone and concerns the delivery of materials, the transfer of the work in process to the next operation or the corridors, or the note that a product is ready to be packed and transferred. If the different actors communicate with the warehouse they usually want to be served directly, so the warehouse often reacts and plans ad hoc. The warehouse needs to deliver information to the production office, copies of letter of contents of the delivered materials and the finished products that are transferred to the customer or the products that are put out. This communication is done by internal post.
5.2 Performance Current Planning Process
To get an indication of the performance of the current planning process, the planning of Tomin Metaal is compared with the framework of Bertrand et al (1998), also known as the BWW framework (figure 5.4). Based on these results possible changes can be made. In the BWW framework the highest decision level is the aggregation level, other possible decision-making at a higher level has no direct consequence for the supply of materials, so are not taken into account in the presented framework. Bertrand et al (1998) assume that influences of overall (umbrella) decision-making will be covered in the aggregation control by logistic parameters.
Figure 5.4 BWW framework to production control and material management (copy of Bertrand et al, 1998).
In this paragraph the performance of the planning process is analyzed to better understand about what happens in and due to planning. The lead time and average work in process are the variables which performance is influenced by the type of order release and planning. Interviews are used to analyze the opinion and current calculation of the performance. A three week observation of the changes in the planning was used as input of the calculations to determine the lead time and the work in process. The conclusions are, if necessary, based at the comparison of the results and theories.
5.2.1 Lead time
How can the lead time be calculated based on the planning data and what is the lead time performance?
There also will be analyzed if the current insight is sufficient and how this information can become more transparent.
The current delivery time is not an absolute known fact. Tomin Metaal produces a lot of different products for many different customers. The lead time is highly variable and hard to determine, but by observing the planning for three weeks it can be concluded that an order has a lead time of about 9 weeks. Only 3 of these weeks are needed for production. The data at which these conclusions are based can be found in table 5.5 and the data that in the measured period on average there were :
114,31 orders in planning each day and, 37,25 orders in production each day
The results of table 5.5 are derived from the second (right) part of the planning (figure 5.1), if the operation is completed or not. Each day is counted how many of the specific operations were completed. In last column the number of new orders in planning was counted each day. As can be seen in the second last column in the measured period it only happens once that the delivery time was adjusted. An order is adapted in the planning at the moment that the offer becomes an order. At the moment that all materials and drawings are complete the order is released to the metal machinery. The measured period in table 5.5 can be seen as an average period, so the results give a representative indication of the lead time of orders at Tomin Metaal.
From the data of table 5.5, it can be concluded that by average 2,47 orders a day are finished, which results in almost 12,5 finished orders a week.
The average lead time can be calculated with the use of Little’s Law, W/λ = L. λ is the average number of orders finished a day (5.5, tenth column), W is the average number of orders in production. So Lproduction = 37,25 / 2,47 = 15,08 workdays, which means 3
weeks. The average number of orders in planning each day is used to calculate the total average lead time.
Tables 5.5 Summery of planning data
Another conclusion is that in the observed three weeks on average more new orders arrived each day then that there were finished (5.5, tenth & twelfth column). This can, based on the way of planning at Tomin Metaal, lead to a higher amount of work in process. If Tomin Metaal wants to keep their work in process constantly, it can be advised to realize an average lead time of 2 week. (W/λ = 37,25/ 3,73 = 9.99 working days)
Continuity of the order
For the transparency it is important to know if the production is continuous or if the buffer is used as storage between production stages.
To get an indication about the continuity of the lead time, a random sample is taken at the mechanical engineering. This department is observed because the production chief admits that he tries to optimize the occupation degree of the punching machine, even if there is no capacity at the other needed operations to finish the order directly.
For two weeks the orders not in operation where noted. In total 26 different orders were stored at the department and the average waiting time was over 3 days. In this period production was at an average level.
Because these measurements were done in the same period as the measurement for the data from table 5.5, both can be combined. These results can be found in table 5.6. This table must give an indication about the value of the 26 orders.
The data of the waiting orders were derived from a period of two weeks, which are 10 workdays. The data from table 5.5 covers 15 days. In table 5.6 the proportions are compared and it can be concluded that 93% of the orders at the mechanical engineering department have to wait about 3 days before the next operation is started.
Ware house Pre-treatment Mechanical engineering Folding machine Drill shop Welding Out-sourcing
A reason for this could be that the output in number of orders at the punching machine is more then double then at the folding machine (5.5).
Days Finished orders Waiting orders %
15 42
10 28 26 92,9%
1 2,8 2,6 92,9%
Table 5.6
Although the results of table 5.6 are not significant, but only a snap-shot of the situation, it can be concluded that most of the orders do not have a continuous flow. There are possibilities for improvements left. This in confirmed by the number of orders waiting at the corridors to get welt, see figure 5.7.
Figure 5.7 Picture corridors and mechanical engineering department; orders waiting for welding
5.2.2 Work in Process
As described in the conceptual model the amount of WIP has a central position. In this paragraph the current performance of the amount of WIP is being described based on the sub question:
What is the average amount of work in process and what can be concluded from this performance?
Current amount of WIP
The first indications of the amount of WIP were based on the calculations made on the planning schedules. In this planning the different departments were split up, only the punching machine is separated. The control of the planning is not kept up consistently; especially for the separated punching machine this is important. Due to this the orders waiting for the punching machine are included in the metal machinery (figure 5.8). Based on the data at the planning and the number of operators available, the amount of WIP has a maximum of 3,5 orders waiting per operation.
Despite of these results, the chief of the production office is convinced that especially TIG welding, but sometimes also the punching machine, the sawmill or the folding machine have more orders waiting per operator. Therefore a more specific measurement is needed. In the planning it is also hard to measure that not all operations can be performed with just one operator. This is taken into account with new measurements.
To get a reliable impression of the amount of WIP at the different workstations samples were taken. At the plan board at the department the orders are divided between the Sawmill/Cutting, Mechanical engineering, Drilling and Welding department. During the observation the number of operators was counted. The period in which the sample was taken was busy, but representative, there was noted that the TIG welding was behind for approximately 2 or 3 weeks.
On plan board Waiting orders n operators per operator per efficient operator Sawmill/Cutting 26 2 13 26 Metal machinery 37 7 5,3 5,3 Drilling department 6 2 3 3 Welding department 27 6 4,5 5,4
Figure 5.9 Orders waiting per operator, observed
Besides the data from figure 5.9, the waiting orders for the starting operations and the TIG welding process were counted. This has led to the results of table 5.10.
Starting operations Number of orders waiting Number of machines Average number of orders waiting Punching: 0 2 0 Sawmill: 16 1 16 Cutting: 6 1 6 TIG welding 13 1 13 Figure 5.10 Orders waiting per operator, observed
Figures 5.9 and 5.10 both show that the number of orders waiting is not consistent. In the measured period the number of orders waiting at the punching machine was minimal.
Based on interviews and observations the causes for the high amount of WIP can be defined:
- No sufficient capacity of operators, this can be explained in several ways. The available operators do not have enough output, or are not able to fulfill the needed operations. For example, there is only one person who is able to do the TIG welding operation. This can also be explained as not having enough employees.
- Only complete a part of an order and performing only a few operations of the other part. If a lot of orders have to be finished in a short time there is sometimes chosen to first finish and deliver a few finished products of the total order, the remaining part is stopped at the moment the chief of production or the supervisor beliefs that the set-up time is short enough. These decisions are based on the importance of the customer.
Although the needed capacity of the orders is not registered, the current measured amount of WIP also gives reasons to reconsider the current planning, because at the moment there is no transparency about the amount of WIP. So there are no agreements on the number of orders waiting for each operation, but it is also not clear to everyone when which order has to be done. The operators wait till instructions of their supervisor or the chief of production.
5.3 Theoretical framework
The production of Tomin Metaal has a high variety of routings, so the production environment of Tomin Metaal can be characterized as a job shop (Bertrand, Wortmann & Wijngaard, 1998). Groover (2001) characterizes a job shop also based on production quantity; up to a quantity of hundred pieces per order can be characterized as a job shop. As can be seen in figure 5.1 (fourth column) the number of products for each order is often larger then hundred. So a part of the production of Tomin Metaal can be approached as a job shop in this report, also because job shop production can easier adapt to customer needs. The other part, the large batches, can be approached as a batch production structure. These orders often have a repetitive character, which makes it suitable for flow or line production. The process layout used by Tomin Metaal is suitable for all types of productions (figure 5.11).
Figure 5.11 ‘Types of facilities and layouts used for different levels of production quantity and product variety. (Groover, 2001)
can easily flow through the river; the rocks at the bottom of the river are disturbances in the flow. This can be disturbance of machines, lack of materials or equipment, failure, but also set-up times or variation in throughput times. The use of buffers of work in process (high water), can deal with the production problems by masking them. This is the classical approach, but according to the JIT approach problems should be solved, so the rocks should be discovered. This can be done by lowering the water or in other words minimize the buffers (Bertrand, Wortmann & Wijngaard, 1998).
Figure 5.12 ‘River of inventory’ 3
Tomin wants to realize a more lean approach and they also want to create more transparency. By lowering the buffers the problems will become clear which means that the organization becomes more transparent. To realize this Tomin Metaal should also drive the planning based on workload of each workstation. At the moment it is not known what the capacity of each workstation is at Tomin Metaal. Because of the large diversity of the performance of each operator per workstation this is a complex task to realize. But it can be advised for Tomin Metaal to start with a capacity planning. Only then Tomin Metaal will be able to introduce another type of planning to realize an efficient transparent work method to lower their buffers.
5.4 Conclusion of the diagnosis
The current way of planning needs improvement to realize more transparency. The lead time is not transparent due to the type of planning and production. Besides, in the current way of registration and production, it costs a lot of effort to determine the length of the lead time, the continuity of the orders and the amount of WIP easily. To lower the amount of WIP the problems in the production process become visible and the lead time reliable which can even be shortened (river of inventory).
At the moment the workload is not geared to the available capacity, because the available capacity is not exactly known. Tomin Metaal should introduce a capacity planning before Tomin Metaal is able to introduce a different type of planning or workload control in order to be able to lower the amount of WIP. At the moment Tomin Metaal has realized a capacity planning and Tomin Metaal still has problems to control their WIP, it can be advised to set up a planning pilot. The bottlenecks in the process are often caused because the production cannot be produced according to schedule due to absent of an employee because of illness, holiday or another reason.
3
By extending the tasks of their operators Tomin Metaal will be more flexible which results in less stagnation of the orders and less WIP. I understand that Tomin Metaal always will be a company with a broad variety of employees and not every operator is able to fulfill all types of tasks, but (in their own words) this should be considered as a challenge.
5.4.1 Conclusion of the current communication performance
At the moment the communication in the organization occurs in a traditional way. As described in chapter 5.1, the current communication flows are overloaded and not structured. A reason for this is the lack of trust in the perceived information, so this information is checked at the source. It also happens that operators receive information from several supervisors or managers. It can be advised to make new arrangements about communication and to conduct this arrangement consistently. This report will not introduce suggestions for more efficient communication, also because a new logistic system will be implemented in October 2008. Due to this system input of information will become compulsory.
A part of communication which will be very important during all implemented changes is that arrangements will not be made in just one conversation. With the arrangements made during meetings it can be advised to use a PDCA-circle (plan, do, check, act). It is important at all times that everyone trusts the system, because otherwise they will not use it and the same communication schedules will occur. If everyone trusts the system, you can be sure that the information needed is accurate. 5.5 Design of a capacity planning
As concluded in paragraph 5.4 the current planning is not clear to everyone. Besides this, the current way of planning is not transparent, because it causes a high amount of work in process. In this paragraph a proposal of a capacity planning is introduced.
For order release it is important that the central planner knows which order is queued (or being processed) at which capacity groups. The planner needs a detailed overview of the shop floor status (Henrich et al., 2004). For detailed information feedback within the workload control concept, any status change on the shop floor has to be registered and transmitted to the central planner in order to get the actual status overview on the shop floor. Capacity groups are usually expressed in time units (Henrich et al., 2004).
A capacity planning should give management insight of the available or needed capacity of the next three months. The second change is the occupation planning which has a range of three weeks. Finally there has to be made a detailed sequence planning based on priority and nesting which has a range of three till five days. The last two planning systems should be done by the chief of production, who is being assisted by the production office for the occupation planning and by the department chief.
operators is determined by the number of available operators. This means that illness and holidays are also taken into account. The capacity of an operator can only be used in one operation. Also the capability of an operator, based on their WSW indication, can be taken into account.
Capacity Planning Month 21,67 = days per month
M = machine O = operator
Period month 1 month 2 month 3 month 4 month 5 month 6 month 7
Operation Capacity (hr) Capacity (hr) Capacity (hr) Capacity (hr) Capacity (hr) Capacity (hr) Capacity (hr) Cutting Available M 151,7 151,7 151,7 151,7 151,7 151,7 151,7 Available O 151,7 Needed Sawing Available M 173,4 173,4 173,4 173,4 173,4 173,4 173,4 Available O 173,4 173,4 173,4 173,4 173,4 173,4 173,4 Needed Punching Available M 346,7 Available O 346,7 Needed Folding Available M 455,1 Available O 455,1 Needed Drilling Available M 1516,9 Available O 1061,8 Needed Welding Available M 1516,9 Available O 1061,8 Needed TIG Welding Available M 151,7 Available O 151,7 Needed
Figure 5.13 Possible lay-out of a capacity planning
Table 5.13 is a possibility to show if the available capacity is sufficient comparing to the needed capacity. If the needed capacity is less than the available capacity that period, everything is in control. Delivery times of orders will be exceeded at the moment that the needed capacity exceeds the available capacity. A capacity planning is not only used to notice exceeding of the capacity, but is also a tool to rearrange production in such a way that all orders can be fulfilled. There are several ways to rearrange the available capacity. Bertrand et al. (1998) introduce two ways of creating capacity flexibility.
- Transfer of operators.
WSW indication of most of the operators, this expanding is not a good idea for Tomin.
In other words this can be explained as extra machine capacity or extra operator capacity. Increasing the machine capacity at Tomin Metaal is in most situations a long term decision and can be made based on a long term capacity planning. Some punch and fold operations can be performed at an excenter press, which makes the available machine capacity flexible. The availability of the operators is relatively easier to realize. The most operators at Tomin Metaal are not multi skilled and operators often prefer to perform the operation which they perform best.
The showed table in figure 5.13 is a result. Besides a capacity planning they need a reliable machine performance and a precise work preparation to determine the needed capacity, update information about the opportunities and the availability of the operators. The showed table is based on a monthly capacity, but if the planning is used for more detail decisions a weekly capacity planning can be used.
Tomin Metaal can use a capacity planning to determine their weaknesses in the available capacity. Due to the insight in over- and undercapacity Tomin Metaal is able to rearrange the planning, the occupation of the machines or the position of the operators. This is the first step to realize a lower amount of WIP.
5.6 Available planning theory
In theory, besides the capacity planning, there are two parts of the planning, a job pool (pre-shop pool) and shop floor (production) planning (Land, 2006). To realize a new type of planning based on lean principles there are a few possibilities, for each possibility the consequences for Tomin Metaal will be described.
5.6.1 Workload control (WLC)
The first workload control concept is based on the principles of input/output control as defined by Plossl and Wight in 1973 (Henrich, Land, Gaalman & Van der Zee, 2004). As Henrich et al. (2004) explain an order is collected in a so called ‘order pool’ before the order enters the shop floor. A central planner releases these orders periodically to the shop floor in such a way that the workload in front of the capacity groups on the floor will be balanced.
The workload principle has as main goal to control the lead time, but to control this at a department implicates to control the work in process (Bertrand et al., 1998). Bertrand et al. (1989) conclude that the change in workload shows if the company is able to keep the workload steady, which gives an indication of delivery reliability. 5.6.2 Kanban
2006). Working with a card driven material-control-system is not impossible as they say, but in those situations they advise the Polca system.
5.6.3 POLCA
The POLCA (Paired-cell Overlapping Loops of Cards with Autorization) principle is one of the variants of Kanban, the difference is that Kanban gives an assignment to the supplying cell to produce an order and the POLCA system only informes that a new order can be received.
According to Suri (1998; Fernandes & Do Carmo Silva, 2006), POLCA is a mechanism that overcomes the limitations of pull systems in high-variety and/or customization product environments, demonstrating its apparent relevance for the make to order and engineering to order sector. This idea is based on the principles used by Sperman & Zazanis (1992). Pieffers & Riezebos (2006) confirm the definition, but add that POLCA uses route fixed cards, the so-called Polca’s. They conclude that POLCA keeps the lead-time low without suffering of efficiency.
To make it possible to implement a POLCA system, a company must split their production system in segments (cells). This can be an already existing production group, split a production group or sometimes functional structured companies should change to material control systems before it is possible to implement POLCA (Pieffers & Riezebos, 2006).
5.6.4 CONWIP
CONWIP is short for constant work in process; this principle is based on the use of circulating cards to control the workload at the shop floor. CONWIP is a predecessor of POLCA which controls also the workload at ‘singular’ workstations with circulating cards4.
In the CONWIP mechanism, when a unit of material is consumed from the finished goods inventory at the last stage of the production system, a new unit is released into the system (Fernandes & Do Carmo Silva, 2006). They also explain that under CONWIP inventory planned levels are established for the whole production system on the basis of the number of CONWIP cards. Once released into the system, jobs are pushed through work centers together with CONWIP cards, to ensure as much as possible, that the system bottleneck is kept busy. Spearman, Woodruff & Hopp (1990) introduce CONWIP as a system which appears to share the benefits of Kanban (e.g. shorter flow times and reduced inventory levels) while being applicable to a wider variety of production environments. CONWIP production cards are assigned to the production line and are not part number specific.
5.6.5 Lean planning and control system (LPCS)
Slomp, Bokhorst and Germs (2008) introduce a takt time control concept which combines WCL and lean principles and uses the CONWIP principle as planning process. These described theories are the three phases of implementation.
In the case study Slomp, Bokhorst & Germs (2007)5 the three phases to implement the LPCS concept are described as follow. At first nesting the orders was transferred from the project preparation to the shop floor; with the advantage to be flexible till the last
4
http://www.procesverbeteren.nl/POLCA/CONWIP.php
5
moment. Before the second step, implementing CONWIP, can be performed the ‘orderstroom’ has to be made transparent with the use of cards. As described the CONWIP principle is able to keep the amount of work in process continuously. In the third step the takt time principle is introduced, with the use of a FIFO sequence. The introduced takt time principle differs from the well known takt phenomenon in that not every x minutes an order has to leave the shop floor, but on average every x minutes an order has to leave the shop floor. The principle is controlled by the total lead-time, as explained before, the maximum throughput time is WIP (total number of orders at once at the shop floor) multiplied by takt time. By make it visual to the operators if an order has crossed the maximum lead time it will be possible to act on it.
With the use of LPCS Tomin Metaal will be able to have a constant, and as small as possible, amount of WIP which increases the transparency. Due to the use of the takt-time principle it will be able to combine the job shop and the batch orders. To some orders a flow can be realized and the lead time performance is clear, this increases the transparency. The operators can be involved by making the performance visible, like Slomp et al (2008) did with the use of a board which showed the orders to finish and if they are in time or not. Because Tomin is a WSW-company a very well introduced pilot is very important to find out how the operators will respond at the performance board.
5.6.6 Results planning
Based on the described planning theories it should be concluded which type of system suits Tomin Metaal. Based on the descriptions of all theories it can be concluded that they all are used to realize shorter and more reliable (production) lead times and to create more transparency by keeping the amount of work in process low. This suits the best with the lean principle and creating transparency. A pull system controls work in process and measures throughput against the required demand (Sperman & Zazanis, 1992). So the described systems are all pull systems. In the current planning system at Tomin Metaal the orders are pushed trough the operations. The start of an order has a pull character, because they only work at customer order.
6. Warehouse
In this chapter the focus will be at the warehouse and the corridors which are used to store materials which could not be stored elsewhere. The warehouse is used for receiving and storing materials and sometimes even for storing work in process. Finished products are also wrapped at the warehouse.
At the moment no distinction is made between internal and external logistics.
During the period of the research a few different things changed at the warehouse, it all started with the arrival of a Romney shed (figure 6.1b). The redesign of the arrangements of control of the warehouse and the corridors can be found at the end of this chapter.
In this chapter the role of the warehouse in the second sub question will be analyzed;
“Which role has the stock of materials and activities of the internal logistics on the work in process?”
Arrangements and responsibilities of the
warehouse are described in the first paragraph. The described information is received through observations at the warehouse and interviews with the warehouse chief, the operators at the warehouse and with the production office. Some information was verified with operators of the production floor.
6.1 Responsibilities warehouse
The mixed use of the words warehouse and internal logistics can be confusing, but at Tomin Metaal the employees of the warehouse also take care of the internal logistics. In consistence with the current arrangement the delivered materials will arrive at the warehouse, these products are temporary placed at the receive station where they are checked according to the description of the production office. Afterwards the materials must be labeled and pallet registration should take place. Deviations in stock of materials should be passed to purchase; deviations in materials bought on order are passed to the production office.
At the moment that the order arrives at the shop-floor, the work description including with a pick-list is available. The production supervisor or one of the operators needs a pick-list to get the needed materials at the warehouse. The warehouse supervisor or employee must take care that the pick-list will be administrated. At the pick-list there is a different between on stock and on order. The lists are needed to optimize the material administration.
Finished products
Appointments made by the production office about the transport of the products should be passed on to the warehouse supervisor who will announce it at the planning table.
Because it is often hard to store the inventory materials (bought to order, received to order, bought to price), semi finished products and stock finished products at the storage, storage control is very important to Tomin Metaal. The warehouse is responsible to the regulation of the material flow at the warehouse in order to foresee the production in time and in a correct way with the right materials and sending the received finished products to the customers.
a. b.
c. d.
Figure 6.1 a. Clogged corridors, b. The new Romney shed, c. original warehouse arrangement, d. new warehouse arrangement
6.2 Performance
The first step to create transparency is to analyze the production process; the next step will be to allocate the relations between the processes, to understand and influence value creation (INK, 2002).
In this paragraph available and needed space to store materials will be measured and the appointments made about the storage location will be discussed. Work in process is the basis to the measured performance. This in order to answer the sub question;
How often is the space to store materials not sufficient enough?
In order to know if the storage space is sufficient enough to the current performance, measurements are executed at the corridors. Due to these measurements also the possibilities to reduce the amount of materials at the corridors are analyzed.
