Balancing the injection-molding department
Implementation of Lean Thinking Principles at the injection-molding
department of Muelink and Grol BV
Education: MSc. Technology Management
University of Groningen
Supervisor: prof. dr. ir. J. Slomp
Co-assessor: dr. J.A.C. Bokhorst
Supervisor M&G: Jelle Slagter
Author: Julius Schurer, BSc. BASc.
Student number: 1730967
Preface
Before you lies my master thesis which I have conducted at Muelink & Grol BV for the completion of the study MSc. Technology Management at the University of Groningen. The thesis’ objective is to deal with the disturbing signals at the injection-molding department. I want to use this preface to show my gratitude to the people who supported me during this research.
From the University of Groningen I want to thank my supervisor prof. dr. ir. J. Slomp for providing the opportunity for conducting this research via the Lean Operations Research Center (LO-RC), guidance, and support. Also my gratitude towards my co-assessor, dr. J.A.C. Bokhorst, for his attention during my research.
Subsequently, I want to thank Jelle Slagter and Richard Kramer from Muelink & Grol BV for giving me the opportunity, facilities, and support for conducting my research. For the support of collecting and understanding the data I want to thank Egbert Lammers, Raoul Molog, Edwin de Vries and Martin Westhof. From the injection-molding department I want to thank the production chef, Henri Prins, and all the operators for assisting me during my research and the chats we had.
At last, but not least, I want to thank Barbara for the all confidence and love she gave me. Groningen, May 2011
Management Summary
Muelink & Grol BV wants to improve the business performance with the use of lean thinking principles. From the findings of previous examinations the question arises if lean thinking principles could improve the performance of the injection-molding department. Muelink & Grol BV wants to know how to deal with the present disturbing signals at the injection-molding department. The research question of this master thesis is: How should the Production Planning & Control (PPC) system at the injection-molding department at Muelink & Grol BV be redesigned by using lean thinking principles to deal with the disturbing signals? Planning and control are taken as control variables with respect to dealing with the disturbing signals at the injection-molding department. The introduction of Muelink & Grol BV and the problem statement are described in the first chapter of this master thesis. The disturbing signals are translated into the following challenges:
• The changeover times are an important manufacturing characteristic, which should be taken into account. The new PPC system should be organized in such a way that the changeover time is minimized by optimizing the product sequence, and the batch and inventory sizes.
• The new PPC system may not be distorted by rush orders what leads to unplanned changes. Information in the system must be accurate to minimize the amount of rush orders.
• The overview at the injection-molding department has to be improved with the new PPC system. Operator must know how the process situation is at each moment of time. • It is desirable to get a constant production flow within the injection-molding
department, which will smoothen the utilization of capacity.
The current state of the situation at the injection-molding department is given in the second chapter to give a good understanding of the current situation. In this chapter is stated that Muelink & Grol BV uses a closed-loop materials requirements planning (MRP), called ‘BMpro’, as their production planning & control system (PPC), which operates under a make-to-stock philosophy. The control of this MRP system is based upon a planbook, which contains the data from the master production schedule. From the plan book operators get their information about what has to be produced. Thereafter, the lean thinking principles are described for answering the question how lean thinking principles can be used to change the current PPC system, which will result in dealing with the disrupting signals. Going ‘lean’ means eliminating the seven wastes. The following wastes are present at the injection-molding department and must be eliminated or minimized in the new PPC system in the thought of the lean thinking principles:
- Inventory - Overproduction
- Production of defective products
- Waiting time based upon changeover time - Unnecessary worker movement
Besides the elimination of waste, the new PPC system must have an effective uninterrupted flow and products should be produced based upon a pull system.
or minimized with the new PPC system and will be used as design criteria for the new PPC system. Also the lean thinking principal of creating an effective uninterrupted flow will be used as design criteria. From the current situation perspective the design criteria’s ‘overview’ and ‘unplanned changes and inaccurate information’ are not contradicting with the lean thinking principles and will be used as design criteria.
Design criteria that contradict between the perspectives are from the current situation perspective ‘product scheduling’, ‘batch size’, and ‘demand variation’. From the lean perspective this are three types of wastes: ‘inventory’, ‘overproduction’, and ‘waiting time’. The elimination or minimization of these wastes depends on the manufacturing environment. The main contradiction lies between the push production with MRP and pull production with lean. To cope with all the contradictions, a right mix of MRP and lean concepts has to be chosen.
Thereafter, literature is examined, which deals with the contradictions between the current situation and lean perspective. From the reviewed literature it is made clear that the control variable planning should be changed by the usage of EPE, which is a cyclic planning method based on the lean control method Heijunka. EPE is lean production control method that involves creating a fixed cyclic plan through the levelling of product volume and mix, with a continuous focus on setup reduction. The basis of EPE is to make each cycle of the plan as small as possible by doing as many changeovers as are feasible, in keeping with lean principles.
For the control variable control yields that the lean control method kanban can deal with the present contradictions. A generic kanban is used for the new PPC system, which contains all the information needed for the production of a batch and will be presented to the operators by using a control box, which is based upon a visual representation of the used EPE cycle. The advantages of the EPE cycle at the injection-molding department are:
• The fixed EPE cycle results in a stable and predictable production environment at the injection-molding department. Calmness will spread over the injection-molding department. Consequently, more time and effort can be spent on improvement efforts, such as changeover time reduction.
• A lean flow is created by reducing variability by levelling the production of runners, which contributes to the lean thinking principle of creating an effective uninterrupted flow.
• The calculated produced batch sizes result in stable production throughout the year, because the forecast is taken into account in the calculations.
• The calculated safety stock will buffer variation in demand, in order to discharge the variation in production. Moreover, the calculated safety stock levels are lower than the current inventory levels.
• Most of the batch sizes of the EPE cycle are lower than the batch sizes in the current situation as stated in BMpro. Smaller batch sizes result in shorter production lead times, and lower work-progress inventories.
Table of Contents
1. INTRODUCTION ... 1
1.1 MUELINK & GROL BV... 1
1.1.1MARKET...1 1.1.2ORGANIZATION...1 1.2 PROBLEM STATEMENT... 2 1.2.1 PROBLEM DESCRIPTION...3 1.2.2 RESEARCH OBJECTIVE...7 1.2.3 CONCEPTUAL MODEL...7 1.2.4 RESEARCH QUESTION...8 1.2.5 RESEARCH DESIGN...8 2. DIAGNOSTIC PHASE...10 2.1 LAYOUT...10 2.1.1 MACHINERY...11 2.1.2 ALLOCATION...12 2.2 ORGANIZATION...13 2.3 PPC SYSTEM...14 2.3.1 PLANNING...14
2.3.2 ASSESSMENT OF THE CONTROL VARIABLE ‘PLANNING’...17
2.3.3 CONTROL...18
2.3.4 ASSESSMENT OF THE CONTROL VARIABLE ‘CONTROL’ ...20
2.4 LEAN MANUFACTURING...20
2.4.1 LEAN THINKING PRINCIPLES...21
2.4.2 SEVEN WASTES...23
2.5 SUMMARY...26
3. DESIGN PHASE...27
3.1 CONSISTENCY OF CRITERIA...27
3.1.1 CONTRADICTIONS...27
3.1.2 DIFFERENCES BETWEEN MRP AND LEAN...29
3.1.3 SUMMARY...31
3.2 COMBINATION OF MRP AND LEAN...31
3.3 CHOICE OF DIRECTION...34
3.4 CONTROL VARIABLE ‘PLANNING’...35
3.4.1 HEIJUNKA...35
3.4.2 IMPLEMENTATION STEPS OF HEIJUNKA...36
3.4.3 EVERY PRODUCT EVERY...38
3.4.4 ADVANTAGES OF EPE ...39
3.4 CONTROL VARIABLE ‘CONTROL’ ...40
4. CHANGE PHASE ...43
4.1 CLASSIFICATION OF RUNNERS, REPEATERS AND STRANGERS...43
4.2 DETERMINING THE OPTIMUM PRODUCT SEQUENCE BASED ON CHANGEOVER TIMES...44
4.2.1 MOLD...44
4.2.2 2K PRODUCTS...45
4.2.3 MOLD – MACHINE EFFICIENCY...45
4.2.4 MACHINE CONSTRAINT...46
4.2.5 RAW MATERIAL...46
4.3 CALCULATING THE OPTIMUM BATCH SIZE...47
4.4 VALIDATION OF THE OUTCOME...50
4.4.1 SAFETY STOCK...50
4.4.2 ACCURATE DEMAND INFORMATION...53
4.5 IMPLEMENTATION...53
4.5.1 EXAMPLE OF PRODUCT FAMILY 1...54
4.5.2 FORECAST...58
4.5.3 INTEGRATION OF MRP, EPE, AND KANBAN...59
5. CONCLUSION ...61
6. DISCUSSION & RECOMMENDATIONS ...63
1. Introduction
Within this chapter an introduction of Muelink & Grol BV is given. The products, market, and organization are described for gaining a good understanding of Muelink & Grol BV. Thereafter the problem statement at Muelink & Grol BV is presented.
1.1 Muelink & Grol BV
Muelink & Grol BV is the leading manufacturer in Europe of Flue Gas Systems. Muelink & Grol BV is founded in 1932 and developed over the years an extensive knowledge of legislation in several countries in and outside Europe. Muelink & Grol BV operates a quality system which is certified according to NEN-EN-ISO 9001:2001.
Muelink & Grol BV is seen as a total supplier of flue gas and ventilation systems by its customers. The wide range of products is produced in aluminium, stainless steel and plastic (PP/PPs). As a partner of several leading boiler manufacturers all over Europe Muelink & Grol BV even contributes to solve flue issues inside and outside the boiler. In these cases its technology and knowledge of its skilled engineers are shared with those of its customers. Together with its customer Muelink & Grol BV always tries to find the right and cost efficient flue solution.
1.1.1 Market
The flue gas systems produced by the Muelink & Grol BV can be divided over three distinct markets. The first and biggest market is the central heating market (about 95%). About 3% of the products are produced for the industrial heating market. The remaining 2% are products for the ambiance lightning market. The distribution of these percentages is somewhat different for Muelink & Grol BV. About 85% are produced for central heating market, 10% of the produced products are produced for the industrial heating market and the remaining 5% are products for the ambiance lightning market.
The Muelink and Grol group, together with the three biggest competitors holds about 90% of the market. These competitors are Cox Geelen, Ubbink and Centroterm. The customers of the Muelink and Grol group are Original Equipment Manufacturers (OEM’s), which produce end products. The biggest customers for the central heating market are the following boiler manufactures: Remeha, Vaillant and Bosch. The demand for plastic inner tubes in this market is increasing corresponding to the metal inner tubes. Because the temperature of the flue gas is decreased (120°C) by high efficiency boilers, the use of cheaper plastics is possible.
1.1.2 Organization
Figure 1: M&G group production location around the world 1. Muelink & Grol BV 2. Burgerhout BV
3. Slagter BV 4. interActive BV
5. Anjo BV 6. Altecnic Ltd
7. sa Isoleco ny 8. Cheminées Sécurité 9. Ant Kalip 10. Muelink & Grol Italia 11. Dura-Vent
The organizational structure of Muelink & Grol BV can be characterized as functional. Each department of the organization tends to perform a specialized set of tasks. Two directors run Muelink & Grol BV. The general director is responsible for the primary production process and the commercial director for sales. The commercial director is head of: external sales, customer service, engineering, and R&D. The engineering department is divided in sales and design. Sales engineering focuses on custom designed systems and design engineering focuses on standard export systems. The general director is head of purchasing, operations, and the business office. The operations manager directs production, process engineering and the technical service. Production manages the four different production departments including the injection-molding department
1.2 Problem statement
Besides the finding of the previous examinations, the question of improving the performance of the molding department is triggered by the decrease in efficiency at the injection-molding department (see Figure 2). Efficiency, calculated by Muelink & Grol BV as planned machine production hours divided by actually used machine production hours, is a business performance indicator at the injection-molding department. Efficiency higher than 100 % can be clarified by a too high chosen value for planned machine production hours.
Figure 2: Efficiency injection-molding department in production year 2010 1.2.1 Problem description
The new management at Muelink & Grol BV reduced inventory levels and batch sizes with the thought: “Costs have to be reduced and the most simple way to achieve this is reducing the level of inventory and batch sizes. When this will result in problems at the shop floor we will put it back into the original situation.” This is a sort of ‘beep’ method. When there sounds not a beep from the shop floor anything will be fine and if it happens the situation will be examined.
Despite of the disturbing signals from the injection-molding shop floor nothing has changed yet. For addressing these signals, interviews were held with most of the injection-molding employees including the production manager and production chef. The signals are divided into the following challenges which are present at the injection-molding department:
Product variety
At the injection-molding department a large variety of products are produced which results in a lot of changeovers. Many changeovers results in a loss of capacity at the injection-molding department.
Production sequence
sequence can be adjusted to this. For instance, placing two products from the same family behind each other in the sequence will utilize less changeover time than a placing two products behind each other from different product families. Making the decisions about the production sequence is the task and responsibility of the planning department and not of shop floor.
From the interview it became clear that the planning department has not always a good feeling about the shop floor. They do not known what the effect is of certain actions they make at the shop floor. For giving a real-life example about the raising questions towards product scheduling a situation on 12-1-2011 will be given where an operator deviated from the actual production scheduled.
After the production of product number 96992 the scheduled production sequence indicated that product number 97005 should first be produced and thereafter product number 97095. Product numbers 96992 and 97005 are produced from the same mold (#1900) and therefore it is a good scheduled sequence with observance of product families. However, this sequence was not praised by the operator due to his other angle of incidence based upon the changeover working method. He explained: “The injection-molding machine will resume faster with production and therefore obtaining a higher machine utilization when we changeover to product number 97095, because it is a changeover with a different mold and can directly be placed into the injection-machine and will resume production. If you look towards the planned changeover I have to remove the mold from the injection-molding machine, change it over for the other product number, and place it back in the injection-molding machine. During this changeover the injection-molding machine was idle, which is a waste of valuable time”.
This example is in contradiction with the description about scheduling product families in sequence. It becomes clear that not only planning, based upon product families, is of influence but also how the control is organized. However, for certain product families a changeover within the mold is faster than changing towards another mold. This yields for instance for the changeover between articles 93386 and 93396, which will be presented later in this thesis. This difference in the reduction of changeover time should be taken into account. Concluding, the production sequence is important for gaining a reduction of the changeover time.
Besides the reduction of the number of changeover and the times it takes there should be looked at the moment when these changeovers occur. It cannot be right that changeovers have to be performed simultaneously, assuming that only two changeovers can be performed at the time due to operator capacity.
Batch size
level of 4000, and a batch size of 3000 that will be increased towards 5000. Now you have a situation that a products batch size is higher than its maximum inventory level. This job order will not be placed unless the maximum inventory level is made higher so the batch can be stored.
Unplanned changes and inaccurate information
The situation, as intended by the planning department, is not always present at the shop floor. It often occurs that injection-molding machines are ahead or behind their schedule. Rush orders have their contribution on this. They disrupt the existing schedule, which results in an unplanned change. Rush orders are planned due to inaccurate information in the Material Requirements Planning (MRP) system, namely the level of inventory is not always correct. The actual level is not always in line with the theoretical level, which results in a rush order that will level the inventory.
Overview
The overview of the injection-molding department is lost. Overview is needed to have a better controlled process. Operators find it hard to say what the current situation is at a particular moment in time. It is not always clear which products are running on a machine and how much is made already. This lack of overview is partly caused by the restriction of reporting the production orders ready and the way planned production is presented. This reporting is only possible during 14:30 - 7:30, because the employees at the injection-molding department may not mutate within these hours due to a restriction of BMpro (see section 2.3.3 ‘Control’). Thus the completed production orders within the given mutation time restriction will not be seen by BMpro as completed, which results in a decrease in the overview. Also the way planned production is presented contributes towards a decrease in overview. The planned production must be reviewed from a plan book, which is hard to read based upon its overview structure, and the data in the plan book is not always up to date.
Some overview is lost due to shift changes. The current state of the process is not always right communicated towards the employees in the new shift. Next to the shift changes, each shift has their own sort of work method, which results in the absence of a general work structure. Demand variation
Figure 3: Capacity utilization 2009
Figure 4: Capacity utilization 2010
When inspecting figures 3 and 4, you can see a trend/similarity located at the beginning and ending of a production year. The utilization of capacity at these periods is high due to a high demand. In de middle of the production years there is a lower demand for capacity.
In the past, the injection-molding department was fighting this variation in demand by producing more than needed during the quiet periods of the production year. This overproduction was used during the busy periods to give the shop floor some ‘air to breath’. But the new management reduced this overproducing, because they have decreased the inventory levels with the mindset to reduce costs.
products are produced, which can be translated into an increasing amount of used production hours to complete a batch.
Now the management of Muelink & Grol BV wants to intervene. They do not think that it is the right solution to go back to the old situation with higher inventory levels and higher batch sizes but to investigate how to deal with the disturbing signals at the injection-molding department.
1.2.2 Research objective
Based upon the problem situation the research objective is stated as: Dealing with the disturbing signals at the injection-molding department 1.2.3 Conceptual model
In order to identify the problem of dealing with the disturbing signals the injection-molding department a conceptual model is made.
Figure 5: Conceptual model
usefulness of this control variable. Planning is an important variable according to Anupindi et al. (2006). They state: “Process planning involves structuring the process capability, flexibility, capacity, and cost efficiency. The long-run goal of process planning is to produce and deliver products that satisfy targeted customer needs.”
Combing the two control variables will give one main control variable: the Production Planning and Control (PPC) system. The aim of production planning and control is producing the right part, at the right time and at a competitive cost (Spearman et al. 1990). Functions of a PPC system include planning material requirements, demand management, capacity planning and the scheduling and sequencing of jobs (Stevenson et al. 2005). For the injection-molding department an optimal PPC system has to be selected from the available systems (Starbek and Grum, 2000), which will deal with the disturbing signals at the injection-molding department. Olsson and Johansson show that several researches have stressed the need for identifying key characteristics that affect the design of the PPC system, and indicate that the design of the PPC system depends on market, product and manufacturing characteristics. The mentioned disturbing signals in section 1.2.1 ‘Problem description’ can be divided under these characteristics and be used as design criteria for the new PPC system. The control variables planning and control will be adjusted with lean thinking principles to deal with the disturbing signals at the injection-molding department. According to Slomp et al. (2009) lean production control principles are a popular way of improving performance of a production system. The utilization of lean thinking principles is the right choice due to the consistency between the new managements philosophy and lean philosophy, which are both based upon reducing inventory levels, batch sizes, and costs.
1.2.4 Research question
Based on the research objective and conceptual model a research question is formulated: How should the Production Planning & Control (PPC) system at the injection-molding department at Muelink & Grol BV be redesigned by using lean thinking principles to deal
with the disturbing signals? 1.2.5 Research design
This thesis is divided into three phases of research based upon the DOV-model of de Leeuw (2003). The first phase represents the ‘D’ in the model of de Leeuw (2003), which is the diagnostic phase. The current situation at the injection-molding department is described within the diagnostic phase to indicate the cause of the disturbing signals. This result will be used in combination with the lean thinking principles to describe the design criteria’s for the new PPC system that will deal with the disturbing signals. Within the diagnostic phase the following sub-questions will be answered:
- How is planning & control in the current situation organized? - What is the affect of planning & control on the disturbing signals? - How do lean thinking principles deal with the disturbing signals?
conducted for obtaining proper solutions for redesigning the current PPC system. Within the design phase the following sub-questions will be answered:
- How do lean thinking principles redesign the current PPC system to deal with the disturbing signals?
The third phase represents the ‘V’ in the model of de Leeuw (2003), which is the change phase. From the design phase is made clear how planning and control must be filled in for the new PPC system to deal with the disturbing signals. The new PPC system is given in the change phase and a theoretical implementation is conducted. The future state of the injection-molding department is described. The third phase will give answer the following sub-questions:
2. Diagnostic phase
This chapter describes the current state of the injection-molding departments planning and control, which are of importance on the disturbing signals. By reviewing the current state it will be clear how the injection-department runs and will help to address some issues/problems, which influence the disturbing signals at the injection-molding department. Before describing the current state of planning and control the process of injection-molding will be described to given a good understanding of the process which is needed for further reading of this diagnostic research thesis. For the visual rendition of the current situation the control variable ‘design’ and ‘organization’ will be described but it will not further impact this thesis.
Injection molding is a manufacturing process for producing parts from both thermoplastic and thermosetting plastic materials. Plastic granules are fed to the machine through the hopper and enter the heated injection barrel. The molded plastic granulates are injected by a reciprocating screw into the mold cavity where it cools and hardens to the configuration of the mold cavity.
Figure 6: Overview of an injection-molding machine
2.1 Layout
The injection-molding department consists of 101 injection-molding machines that are numbered with a plan group number. Each injection-molding machine needs a cooler because the injection-molding process needs cooling to eliminate the created heat during the heating of the raw material. These coolers are placed next to the injection-molding machines and some injection-molding machines are sharing a cooler. A finished product will leave the injection-molding machine by a conveyor belt, which ends at a collector box. The product in the collector box will thereafter be placed in a storage cage. The computer, placed near injection-molding machine number 21, prints the needed labels for the storage cage with product and storage information.
1 Injection-molding machine 20 will not be analyzed during this research due to its dedicated utilization.
More than only the machine is needed for the production of an order, namely a product specific mold and raw material. The molds are stored in a designated area. The raw material consists of two components, namely the basic plastic and pigment. The main material is stored in bins from which the material is sucked into the injection-molding machine through tubes. The pigment is transported manually by the operator into the injection-molding machine.
Figure 7: Layout of the injection-molding department 2.1.1 Machinery
The injection-molding department consist of 10 various injection-molding machines and 9 of these will be used in this thesis. The injection-molding machines differ in tonnage and the ability to make ‘two-component’ products. A two-component product is made from 2 different raw materials and therefore two separated hoppers are needed on the injection-molding machine.
Plan group number Tonnage Two component
4 200 √ 5 160 6 200 √ 7 200 √ 19 100 21 200 22 160 23 250 24 270
2.1.2 Allocation
The injection-molding machines are not identical to each other, because each machine has its own capability and restriction. Not each mold (product specific) can be placed on a machine due to the machines capability and the design restriction of the fixed hoist crane because a mold cannot be placed when its weights more than the maximum lifting power of the hoist crane.
For providing a visual rendition of the current situation manufacturing characteristics the allocation of products towards the injection-molding machines is given. Before analyzing the allocations of the products on the injection-molding machines a Pareto analysis is performed, because the large product variety (i.e. during the production year 2010 there were 344 different products produced) will obscure the visual rendition and overview, and the complexity of data collection / processing.
A Pareto analysis that ranks the sum of the annual processing time of a product has been chosen. The Pareto analysis is not conducted with the annual number of products (volume) because the differences between processing times of the products have to be taken into account. By this way the capacity utilization is taken into consideration.
Figure 8: Pareto analysis of produced articles in production year 2010
For the Pareto analysis a value for demand is used, which is based upon the amount that has been produced at the injection-molding department plus the amount in stock, which was used for demand. Products 93388 and 93389, and products 93390 and 93391, are manufactured simultaneously from one mold. Therefore they are seen as one product and are taken together. Product 95206 is not taken into consideration because the product is currently produced continually on plan group 4 due to high customer demand. Therefore, plan group 4 will not be available for the production of other products.
The allocation of the products on the different injection-molding machines is analysed by using the historical data of production year 2010. With the article-machine matrix an overview is constructed which shows the allocations of the high utilization articles and the current allocation flexibility of the various plan groups (see Figure 9).
Figure 9: Article-Machine overview of high utilization articles
2.2 Organization
Within Muelink & Grol BV departments are numbered with a group number and number 19 has been imposed to the injection-molding department. Within this group the machines are divided into the so-called “planning groups” 4 t/m 7 and 19 t/m 24. The placement of these machines can be looked up in figure X.
The injection-molding department uses a working week of 5 days, starting at Sundays 23:00 hours and ends at Fridays 23:00 hours. Within this working week 3 teams provide the production. Each team consists of 6 workers; the first men and 2 operators are employed by Muelink & Grol BV, and the 2 wrappers and KVT operator are employed by an agency. The injection-molding department consists of four organizational layers, namely production manager, first men, operator, and wrapper.
occupation of workers, which are needed for a given period in time. His direct contact with the shop floor goes through the first men.
First men: Each work team has his own first man who is the controller of the group. If any problem occurs, the first man is the first to solve it. Besides his own work as operator, he is the contact for the other operators and wrappers in his group.
Operator: The operator ensures that the orders are produced. His tasks consist of process monitoring, controlling the wrappers, and transporting the cages.
Wrapper: When a product is produced it must be go through a visual quality check performed by the wrapper. When the product passes this visual control the wrapper will place the product in the cage. Besides packing, the wrapper will undo the product from injection-molding residuals, which are attached on the product after leaving the injection-injection-molding machine.
2.3 PPC system
The Production Planning and Control (PPC) system is in literature also referred as the Manufacturing Planning and Control (MPC) system. The PPC system concerns the planning and control of the production process. It provides information to efficiently manage the flow of materials, effectively utilize people and equipment, coordinate internal activities with those of suppliers, and communicate with customers about market requirements (Vollmann et al., 1997). The PPC system will be analysed by dividing it into the separate sections planning and control for reviewing the current PPC system of the injection-molding department at Muelink & Grol BV.
2.3.1 Planning
Muelink & Grol BV uses a closed-loop materials requirements planning (MRP), called ‘BMpro’, as their production planning & control (PPC) system. This PPC system operates under a make-to-stock (MTS) philosophy where products are produced in anticipation of demand. MRP determines how much of and when dependent-demand items should be ordered to satisfy requirements for an end-item. The MRP computer system determines the required quantities of parts and components and sends out orders for them. The closed-loop MRP is a more complex system for PPC because it considers both materials and capacity requirements. It is called closed-loop MRP because it takes into account feedback about orders to be released to the shop floor.
Prior to releasing production orders to departments, closed-loop MRP uses a separate capacity-planning module to check the feasibility of the orders. The workload imposed by current planned orders is added to the already scheduled workload, and the sum is checked against the available capacity. If the anticipated workload exceeds the capacity limit, a warning is issued. This is a signal that the MPS should be revised or that a way needs to be found to increase capacity.
items procured from outside suppliers, the system issues purchase orders. This MRP procedure is be made visual by the “virtual inventory” menu of BMpro, which is presented in Figure 10.
Figure 10: Virtual inventory menu in BMpro
Determining backwards in time is visualized in Figure 8. The MRP system BMpro determines if the inventory level is sufficient for delivering the order to the customer with the right due-date. This process can be followed in BMpro under the menu ‘virtual inventory’. The inventory level decreases when a order requires items held in this inventory. When an order ensures that the minimum level of inventory is reached or the level of inventory is not sufficient to meet the order, an automatic generated production proposal is send to the planning department. Now it is the planning departments own choice to send a job order to the shop floor for supplementing the inventory level. This choice is driven on the need of the product and is also affected by the position in the production sequence. When a job order is send to the shop floor it size is restricted by its fixed batch size, which is determined for each product.
The major inputs of the MRP procedure are the master production schedule (MPS), the bill of materials (BOM), and the inventory and lead-time information file.
• Master production schedule (MPS)
The MPS specifies the end products to be made, the dates they need to be completed, and the quantities required. The MPS is based on the production plan (demand forecast), existing customer orders, and current information regarding the capacity of the manufacturing facility.
• Bill of materials (BOM)
The BOM is a recipe for a product. It indicates all of the raw materials, parts, components, and assemblies required to manufacture a product.
Figure 12: Bill of msterial menu in BMpro • Inventory and lead time information file
The inventory and lead-time information files contain inventory and lead-time information for all raw material, work-in-progress, and finished goods. The MRP system must know how long it takes for all incoming parts to be purchased and for each manufacturing process to be completed. The system must also know exactly how many units of each purchased part, each manufactured component and assembly, and each finished product exist on the shop floor and in the warehouse.
The major outputs of the MRP procedure are the inventory transactions and production orders & purchase orders.
• Inventory transactions
An inventory transaction occurs whenever the MRP procedure reviews and updates a record. • Production orders & purchase orders
Production and purchase orders are scheduled to satisfy the requirements of all items. This scheduling is done by the planning department with the use of the following ‘planning rules’:
o A production order for a foreign customer will be planned 2 days earlier than the delivery date due to the needed transport time. For a native customer this is 1 day. o Fixed processing times and changeover times are used for making the capacity
planning (capacity requirements). From these data, the daily production is calculated which identifies the maximum production per day.
o The sizes of production orders are constrained by the minimum and maximum inventory level. The batch sizes of the production orders are fixed, whereby the batch size can be calculated by; maximum inventory level minus minimum inventory level. o For the allocation of orders to planning groups an excel query file is used which
contains historical data of made allocations in the past. A plan group with the most allocations of a particular product is seen as the best choice for the manufacturing of this product.
Figure 14: Overview of the MRP system 2.3.2 Assessment of the control variable ‘planning’
long. For instance, processing large batches will negatively affect the flexibility of the injection-molding department, because a large amount of capacity is used for the production, which can not be used for other batches.
The historical data of the made allocations is used for future allocations of new production orders. This way of planning affects the disturbing signal of the production sequence. The new design of the PPC system should take the production sequence into account to minimize the changeover time at the shop floor, because the changeover times are long and affect the disturbing signals at the injection-molding department.
Unplanned changes occur due to the utilized MRP system. The MRP system produces under a make-to-stock policy, where products are produced according to the forecast. A rush order will be planned here when the inventory of a product is not capable of meeting demand due to demand variation, which results in an unplanned change. Lead-time will decrease due to backwards movement in the production sequence, whereby the dependability and the overview of the process decreases. The new design of the PPC system should counter these unplanned changes.
The current planning does not fully take the forecast into consideration, which causes into peaks and valleys of the capacity utilization. The peaks and valleys in the demand have to be smoothened by creating a constant production flow within the injection-molding department. Variation in demand should be absorbed by the use of a constant production flow throughout the whole production year.
2.3.3 Control
Operators perform their work on the basis of a plan book, which contains lists of non-completed work orders generated from the MPS. From the plan book operators get their information about what has to be produced.
Within the plan book non-completed work orders are divided by plan group. When a machine is ready for processing another order, the operator first has to consult the plan book for the next planned order in line. Once this order is found, the operator has to fill in and print out a ‘production order’ form at the office. The production order form contains information, which is needed to complete the injection-molding process successfully.
This information includes among others:
• Plan group number: On each production order several product allocation options are given. The operator has to mark the plan group number where the order will be processed. This allocation information of orders to plan groups can be found in the plan book, which already is determined by the planning department.
• Data: The production data of the injection-molding process are included. For instance, mold number, raw material, software program number for the injection-molding machine, etc.
• Packaging: When an order is produced it is being packed into a cage at the department. The type of cage is given at the production order form.
Before the production of an order can take place, the injection-molding machine must be made ready for production, a changeover has to be performed. Each product uses its own product specific mold and therefore each different production run needs a changeover to occur. The molds are stored in a small warehouse near the injection-molding department. Besides the changeover of the mold, there must be examined if the right raw material for the new order must is present at the shop floor. If the amount of raw material is not sufficient or other raw material is needed it will be ordered from the warehouse X52. Ordering from warehouse X52 can only be done during their working hours, namely between 7:15 am – 4:00 pm.
During the production products leave the injection-molding machine by a conveyor. The conveyor ends at a collector box, where workers perform a visual quality check and dispose any product residual. From the collector box products are packaged into a cage. This cage is transported towards the indicated inventory spot when it is full or the order is fully processed. During the packaging of products into the cage a form is filled in which keeps track of who has done the packaging, the amount of product in the cage, number of defects, and the date and time of packaging,
An order must be reported as ready when it is completed to keep the MRP system up-to-date. This ‘ready’ reporting starts with the operator putting the filled in production order form, plus the packaging form and defective products form (described later), into a mailbox called ‘report ready’. The finished order (filled in production order forms) will be put into the MRP system so the inventory level of the relevant product will be up-to-date. This ready reporting can only be done during particulars part of the day, namely between 7:00 – 8:00 am and 3:00 – 8:00 pm. The restriction between 8:00 am – 3:00 pm is based upon that the business office must be able to mutate during office hours. Reporting orders as ready by the production departments’ results that the business office is excluded from making mutation in the system, whereby they cannot perform their daily tasks. The other restriction between 8:00 pm – 7:00 am is placed due to the night program of BMpro, when BMpro is processing the past days activities.
Next to reporting the finished order, the amount of defective/rejected products and successfully produced products is placed by the operator onto the production order form. The amount of defective parts is also written on a more specific form where a description of the nature of the rejection must be written on.
The amount of defective products is put into the program “Gateway”, which is a quality control program within Muelink & Grol BV. Gateway is not linked with the MRP system and therefore the amount of produced products of an order has to be adjusted in BMpro because the gross production is reported ready.
Figure 15: Overview of control 2.3.4 Assessment of the control variable ‘control’
The current execution of control results in the disturbing signal based upon ‘overview’ mentioned in section ‘1.2.1 Problem description’. Information concerning the next upcoming batch must be collected from the plan book, which is not always up to date and the information is presented by a confusing structure. Using the plan book as guide for the production result in a lack of overview for the operators. Therefore, the new PPC system should be redesigned for creating a better overview.
2.4 Lean manufacturing
At this point are the disrupting signals and current state of the injection-molding department known. Now it is the question how lean thinking principles can be used to change the current state, i.e. control variables planning and control, which will result in dealing with the disrupting signals. In this section lean thinking principles will be explained and will be linked towards the current state at the injection-molding department.
Womack et al. (1990) quotes: “There are two fundamentally different business systems, two ways of thinking about humans work together to create value. One system – mass production – was pioneered by General Motors in the 1920s as it passed Ford to become the world’s largest industrial enterprise. This system was then widely copied and used by enterprises in practically every industry all over the globe. The other business system – lean production – was pioneered by Toyota in the twenty years immediately after World War II and is now rapidly diffusing to every corner of the world.”
right time, and with minimum costs. Products will flow rapidly and smoothly through processes, which is illustrated with an example adopted from Slack and Lewis (2008).
Figure 16: Traditional approach – buffers separate stage (adapted from Slack and Lewis, 2008)
Figure 17: Lean approach – deliveries are made on request (adapted from Slack and Lewis, 2008) Figures 16 and 17 show both a three-stage production process. The traditional approach (Figure 16) has between each stage a buffer so each stage can produce independent from each other. The production at each stage is controlled by a push signal, which pushes the production throughout the production process. With this approach throughput times will be high because the products are temporary stored in the intermediate buffers. The main argument against this traditional approach lies in the very conditions it seeks to promote, namely the insulation of the stages from one another (Slack and Lewis, 2008). When a problem occurs at a stage with the traditional approach the problem will not immediate been seen at another stage due to the intermediate buffers. So, the people at the stage which have a problem are one their own to solve it. However, with a lean approach (Figure 17) the problem would have been directly noticed because there are no intermediate buffers present. The product flow within a lean approach is triggered by the stage ahead when a product is needed. This means that problems at any stage will be seen.
2.4.1 Lean thinking principles
Companies want to maximize their competitiveness and profitability by distinguishing themselves from the competition for generating customers. Next to this, they have to minimize their costs to increase their profits. Minimizing cost should me achieved by eliminating the waste that doesn’t add value. The operational goals of lean thinking are to improve quality, minimize lead times, and reduce inventory levels (Wisner and Stanley, 2008). To accomplish these goals, lean thinkers start with an analysis of their processes. Womack and Jones (2003) made a five-step thought process for guiding the implementation of lean techniques.
1. Specify value
2. Identifying the entire value stream
Identifying the entire value stream for each product is the next step in lean thinking. The producer must evaluate the value stream for each product. This step will expose non-value adding steps (i.e. waste).
3. Create flow
After the exposure of non-value adding steps, an effective uninterrupted flow must be created. 4. Establish pull
Once a flow had been created, products should only be produced based on a pull system. 5. Seek perfection
As organizations begin to accurately specify value, identify the entire value stream, make the value-creating steps flow continuously, and let customers pull value from the enterprise, the process starts all over again for seeking perfection.
2.4.2 Seven wastes
The main objective of lean is the elimination of waste, anything that does not add value to an item or a process. Going ‘lean’ means eliminating the seven wastes. Before eliminating the seven wastes they must be analysed, which can be seen as step number 2 of the five-step thought process given by Womack and Jones (2003). In this section the seven wastes within the field of planning and control at the injection-molding department will be analysed. These wastes should not be present in the new PPC system according to the lean thought. Taiichi Ohno, the main architect of the Toyota Production System, classified seven types of waste in manufacturing (Nicholas, 1998; Anupindi et al., 2006; Bell, 2006).
• Inventory
Companies have overstocked inventories due too much unnecessary inventory in the form of raw materials, work-in-progress, and finished goods. This is often the case when a company uses a MRP system, which pushes production without considering the actual demand. Inventory represents items waiting for something to happen, a waste in that there are costs associated with keeping items waiting and lost time since no value is being added to them. Since the capital needed to produce the items in inventory cannot be invested elsewhere, there is an opportunity cost as well. The funds could have been spent elsewhere and generated a higher return.
At the injection-molding department there is a lot of inventory present due to the MRP system, which utilizes a make-to-stock production philosophy (see section 2.3.1 ‘Planning’). The make-to-stock production philosophy makes products in anticipation of demand. These products go into finished goods stocks before being withdrawn to fill customer orders. Each product of the total assortment of products has a minimum inventory level, which causes a particular inventory level, which is theoretically always present. Besides a minimum inventory level, a maximum inventory level is present. This maximum level controls the amount of inventory for bulging out.
When a batch is being processed a fixed batch size is used for the production. Finished products will be placed in cages until they are full or the batch is fully processed. These products are marked as work in progress (WIP). The size of the WIP depends on the size of a batch, the larger the batch the larger the WIP. When a batch is finished it is reported as ‘ready’ and WIP is from this point changed into finished goods.
The articles cannot be produced without the needed raw materials. Raw materials can be divided into the main material and pigment. These materials are held in stock with a minimum level of inventory.
• Overprocessing
The waste of overprocessing belongs to the field of engineering, because overprocessing occurs when the production process is not made optimal for its particular purpose. Therefore, waste caused by overprocessing will not be examined within this thesis.
• Overproduction
Companies produce sometimes more than they have sold because they want to build inventories or because they want to keep their equipment and facilities running. Overproduction will result in more inventory which is, as discussed above, a waste.
Overproduction can be seen as production to build inventories and production to let the equipment and facilities running. These forms of overproduction are both present at the injection-molding department. There is overproduction to build inventories due to the make-to-stock philosophy (waste ‘inventory’) and overproduction for the keeping the equipment and facilities running due to three causes;
1. Filling a cage to its maximum. Operators see this as efficient production, because the mindset: ‘It is a small effort to fill-up the cage, because the changeover made for this batch has caused a lot of time’.
2. Operator availability. It occurs that an operator is not always available for ending of batch by stopping the injection-molding machine. For instance, when an operator is occupied by performing a changeover at another injection-molding machine he is not able to stop the completed batch at another injection-molding machine. An operator can only be at one place at a time.
3. Operator monitoring. Operators do not have always the overview over the process, which results in a too late discovering of a complete produced batch.
• Production of defective products
When products do not pass the quality check they have to be disposed. Production of defective products results in lost processing time and lost raw material, which causes money. When a defective product comes back from a customer not only used resources are lost but maybe also the customer and potential customers.
At the injection-molding department each batch has a start-up phase, which generates defective products. Besides the start-up defects, there are defects causes by disturbances in the process. The distribution of defects under these two causes is unknown due to insufficient information from the quality program Gateway.
• Unnecessary transportation
Because layout is not in the scope of this thesis, the waste of unnecessary transportation based on layout will not be taken into account. However, the sequence of operations is. Planning influence the sequence of operations through assigning orders to a particularly plan group. In the case of the injection-molding department this affect is minimal due to the placement of injection-molding machines. The difference in placement, based on the production process, is nil and therefore is the sequence of operations not used for the assignment of waste caused by unnecessary transportation and will not be taken into account within this thesis.
• Unnecessary worker movement
Workers in motion must not be confused with working. When a worker is working he or she adds value to the product but when a worker is walking towards a needed tool it is not classified as working but as motion. All the wasteful motions including, searching, selecting, picking up, transporting, loading, repositioning, and unloading should be eliminated.
The waste caused by unnecessary worker movement is mainly controllable with the control variable layout, which is not included in this thesis. However the control variable control has also its influence. Operators often walk to the office to consult the planning in BMpro, because there is not a good overview on the situation. This movement is seen as an unnecessary worker movement. Besides checking the planning, operators have to walk towards the office for printing out the production order lists, which are needed for producing a batch.
• Waiting time Waiting time consist of:
• waiting for orders, parts, or materials
As described in section 2.3.3 ‘Control’, the first men orders the needed raw material for the production of the products. Because this is done before raw material is actually needed it is assumed that there is no waiting time for material.
• waiting for a part to be processed
When an operator has successfully performed a changeover at the injection-molding machine for making it ready for the next order and the particularly batch is being correctly produced, the operator has to wait when the batch is produced. The operator performs more work than only making changeover but this falls within the control variable ‘organization’, which is not included within this thesis. Therefore, the waste caused by ‘waiting for a part to be processed’ will not taken into consideration.
• waiting while a machine is being repaired
2.5 Summary
The mentioned disruptive signals will affect the design of the new PPC system because it represent the challenges (sections 2.3.2 and 2.3.4) at the shop floor and are thus important for the new PPC system.
• The changeover times are an important manufacturing characteristic, which should be taken into account. The new PPC system should be organized in such a way that the changeover time is minimized by optimizing the product sequence, and the batch and inventory sizes.
• The new PPC system may not be distorted by rush orders what leads to unplanned changes. Information in the system must be accurate to minimize the amount of rush orders.
• The overview at the injection-molding department has to be improved with the new PPC system. Operator must know how the process situation is at each moment of time. • It is desirable to get a constant production flow within the injection-molding
department, which will smoothen the utilization of capacity.
Due to the analysis (section 2.4.2 ‘Seven wastes’) of the present wastes it is been made clear that the following wastes are within the scope of research and present at the injection-molding department and therefore must be eliminated or minimized in the new PPC system in the thought of the lean thinking principles:
- Inventory - Overproduction
- Production of defective products
- Waiting time based upon changeover time - Unnecessary worker movement
3. Design phase
The new PPC system must meet the criteria mentioned in section ‘2.5 Summary’, which are obtained from the current situation to remedy existing disrupting signals, and the usage of lean thinking principles. It is assumed that when modelling around these criteria’s the disturbing signals at the injection-molding department will be dealt with. In this chapter literature is examined for obtaining the proper completion of the control variables planning and control at the injection-molding department based upon lean thinking principles.
3.1 Consistency of criteria
In this section will examined whether these criteria go along with each other or are contradictory to each others for designing the new PPC system.
The case at the injection-molding department can be seen as a situation where lean thinking principles have to deal with the disturbing signals of the existing PPC system. These lean thinking principles must not be totally out of line with the current situation, which will be analysed within this section. The current situation is a sort of filter for the ‘path of lean’ towards a new PPC system (see Figure 19). Lean thinking principles at the beginning must pass this filter and it is the question which of these principles will reach the end of this path. These remaining principles will be used as design criteria for the new PPC system.
Figure 19: The path towards lean 3.1.1 Contradictions
From the perspective of lean thinking principles it was stated that five wastes were present at the injection-molding department and these wastes had to be eliminated from the new PPC system. Next to the elimination of wastes, the new PPC system should have an effective uninterrupted flow and must control its production process through a pull system.
times. The amount of changeovers and changeover time can be contributed towards the waste ‘waiting time’ where both perspectives want to minimize the amount of time used for changeovers. However, the manner how this must be realized differs between the perspectives. Viewed from lean perspective, the working method concerning the changeover should be changed to reduce changeover time whereby more changeovers could be performed within the same time. Searching for opportunities to reduce the changeover time could do this. The Single Minute Exchange of Die (SMED) method should organize this search for changeover time reduction (van der Meer and Schurer, 2010). Viewed from the current situation perspective, the amount of changeovers and changeover time can be reduced by changing the product sequence made by the planning department and by increasing the batch sizes and inventory levels. However, increasing the batch size and inventory levels is in contradiction with the lean perspective of eliminating wastes (i.e. inventory and overproduction). The new PPC system should make an informed decision to cope with above-mentioned contradiction. Also Crama et al. (2001) discuss above contradiction and speak of the dividing line between planning and scheduling issues which is not always drawn unambiguously in the production management literature. They say the following: “It is usually agreed, in particular, that the distinction between planning and scheduling should not be based only on the horizon length, but should also account for production conditions. For instance, if a production run takes many days, as is frequently the case in process industries, then the scheduling problem may be defined on a horizon of more than one week. Similarly, when setups are time-consuming, lot sizing issues may become crucial and cannot always be deferred to short-term planning models”.
According to the earlier given description of waste, companies have overstocked inventories due too much unnecessary inventory in the form of raw materials, work-in-progress, and finished goods. This is often the case when a company uses a MRP system, which pushes production without considering the actual demand. Muelink & Grol BV uses this MRP system, which utilizes a make-to-stock production philosophy. Inventory here is very important and have to be present at given levels to run the company. This contradiction is based on the differences of the production method between MRP and lean, namely push and pull.
The problem of the distorted planning with a push system caused by rush orders is not present when using a pull system, because the actual demand is considered. The MRP system computes production schedules based on forecast. Once produced, subassemblies are pushed to a next level whether needed or not. If information in the system is not accurate, which is a problem at the current situation, the level of goods needed at this next level is not present and a rush order is released for completing that particularly product. Rush orders are not present with a pull system due its mechanism. The basic mechanism is that production at one level only happens when initiated by a request at higher level. Products are pulled through the system by request.
situation at the injection-molding department. Mirsky (1993) confirms this. He argues that seasonality effects result in demand peaks, which exceed capacity. Thus, planning is necessary to smooth production runs, contrary to the underlying philosophy of pull systems. For smoothing the variation in demand yields that inventory levels must be increased but this is in contradiction with the wastes (i.e. inventory and overproduction) within the lean philosophy.
Concluding, the current situation runs in some areas counter to lean thinking. The main contradiction lies in the difference between the present push system and the future pull system. The lean thinking principles cannot be simply implemented due to the characteristics of the current situation.
3.1.2 Differences between MRP and Lean
Due to the contradiction between the design criteria’s, it seems that it would not be easy to apply all the lean thinking principles at the injection-molding department. In this section more literature is reviewed involving this contradiction of design criteria’s. The discussed contradictions have to be solved by choosing a side or by sitting in the middle of these contradictions, where a sort of break-even point is determined for making a trade-off. There has to be made a choice between the contradicting design criteria. Several researchers have addressed this topic of contradicting design criteria’s, which occur during the transformation towards a pull system while having a push system and are described next.
Dixon (2004) begins in his paper, about the consistency of lean techniques and IT solutions, with a story that illustrates the problem from having a MRP system and going to a ‘lean system’. The story goes as:
There is a story about the owner of a manufacturing firm who went to Japan to learn about lean. He was shown lean techniques in action and was duly impressed with the results: extraordinary productivity levels, scant inventories, very visible control of quality, throughput at light speed, and great customer service. “This is for me,” he said. “But tell me how this ties in with my MRP system.” It doesn’t,” his host replied. “Go home and turn it off.”
The reaction of Dixon (2004) on this story is: “It is doubtful that the owner actually scrapped his manufacturing system, but the story illustrates the dilemma faced by companies with legacy manufacturing resources or enterprise resources planning systems that are also in the throes of a lean journey”. This also yields for the injection-molding department. The present MRP system cannot simply be scrapped due to its make-to-stock environment.
