Production Planning & Control
at
Bollegraaf Recycling Machinery:
Determining the optimal utilization
of newly acquired machinery
I. PREFACE
That which we persist in doing becomes easier, not that the task itself has become easier, but that our ability to perform it has improved.
Ralph Waldo Emerson (1803-1882)
This thesis is the final step of the master Operations & Supply Chains, and with that I finish my study Business Administration at the University of Groningen. Not only do I complete my master with this project, it also marks the end of 20 years of education.
I performed my graduation project at Bollegraaf Recycling Machinery in Appingedam, and I must say that it has been a great time. I really enjoyed my colleagues at the work preparation department, who all have a great sense of humor, but also have been very helpful during my stay. Furthermore I had the honor to compete in, and win, the prestigious Bollegraaf indoor soccer tournament. Besides all the fun, and most importantly, it has been a very educational time.
I would like to thank Mark Boelens for offering me the graduation spot at Bollegraaf and supporting me through the first part of it. Unfortunately Mark found an other job, but I have to say that his successor, and my new supervisor, Bram Bos also aided me a lot during the project, for which I would like to thank him as well.
Furthermore I would like to thank my supervisors from the University, Wout van Wezel and Jannes Slomp for their useful feedback. I would also like to thank Martin Land for the useful insights he gave me during my graduation project.
Finally I would like to thank my girlfriend Elske, my friends and family for their great support and the joy they bring in my life.
II. MANAGEMENT SUMMARY
Due to the changing market conditions Bollegraaf Recycling Machinery (BRM) has decided to expand the capacity at the cutting machine by purchasing two new machines. BRM wants to know how this capacity can be utilized most efficiently. In the current situation BRM was only able to cut the parts of balers in batches of four. The new machines enable BRM to cut in smaller batches, but also have the option to cut the frames of sliding belts, which are currently outsourced. Therefore the following research question was formulated:
“What is the optimal batch size of the new cutting machines with regard to balers and what will be the influence of the additional cutting for sliding belts?”
What is the optimal batch size for balers?
To give an answer to this question first the different production planning and control characteristics, as well as the main performance objectives have been analyzed. The most important characteristics with regard to the optimal batch size are that BRM is a Make-to-Order (MTO) organization, demand is lumpy, and capacity is very flexible. The most appropriate production strategy is therefore a chase demand strategy according to Olhager et al. (2001). The most important performance objectives with regard to the new machines are speed and cost.
By analyzing the changes in the production process it could be seen that the total time at cutting is reduced to a level of 63% of the old operation times. Besides that less deburring and grinding is required in the new situation. Other important considerations were that the time it takes to cut a batch of two and to perform the setup is the same as for a batch of one, and that transport can be combined in case of a batch of two, meaning that, labor, setup and, ordering costs are lower in this case. However, due to the lumpy demand figures, carrying cost can become relatively high. The optimal batch is therefore a trade-off between these aspects. The following three batch size options were considered:
To determine the difference between the total costs associated with each option three production plans were created, each using one of the above options. These production plans were based on the MPS, which was based on the historical sales data of the last half year.
Based on these calculations it could be seen that the third option performs best, however the difference between the three options is not very large. By choosing the third option on average approximately €1.400 can be saved each month over the first option, and €800 over the second option. The economies of scale resulting from cutting in a batch of two do add up to the higher carrying cost of these options. By performing cutting in a batch of two on average €3.900 could be saved over a batch of one each month. As long as the carrying cost of the second and third option stay below this level, meaning that they are not placed in storage for a too long time, they are more beneficial.
It is therefore not beneficial to produce each type of baler using the third approach. For the balers with a very low demand, like e.g. the HBC60 or HBC180, it is too risky to always produce in a batch of two, due to the high risk of long storage duration, and with it high carrying cost. For these types of balers it is in most cases more beneficial to use a batch size of one. In the case two of these ‘low demand’ types have been sold, and the time between both delivery dates does not exceed a couple of months, the third approach will be more beneficial.
For those balers with a higher, more stable demand, the third batch size option should be used, since the risk of high storage duration is lower for these types. Based on the historical demand figures of the last two year, the following types could be produced using the third batch size policy: HBC80, HBC120, HBC120K, HBC110, HBC110K, and the HBC140. The other types should initially only be produced in a batch of one. In the case two of these ‘low demand’ types have been sold, and the time between both delivery dates does not exceed a couple of months, the third approach will be more beneficial for these types too.
required at assembly. This advantage is however negated by the fact that certain customer specific parts have lead times of eight weeks. This means that even though the frame is in stock, it still would take eight weeks before assembly can start. Since it takes up to a maximum of 48 days to produce a frame at the ‘Stadsweg’, excluding cutting but including spraying and painting it at Smit Coating, there is no additional benefit of using the second approach over the third approach. Therefore the third approach is also the most suitable with regard to the speed objective.
Besides that does the third approach also lead to more flexibility than the first approach. Since half the hours of load are required to produce two balers, capacity utilization will be lowered, meaning that more capacity is available to produce the frames of sliding belts.
What will be the influence of the cutting of sliding belts?
To get an answer to this question we performed a rough cut capacity planning, considering the demand for the last half year. This RCCP contained both the load of balers, as well as the load of sliding belts. The average utilization lies at 78,60%, which shows that in total sufficient capacity is available to insource all sliding belts. However there do occur substantial overloads in certain weeks. In the case of balers there is probably sufficient flexible capacity later on in the process to cope with this, however for sliding belts this is a lot harder due to the short lead times. Therefore possible actions to cope with this overload have been identified, these were the following:
• Raising capacity by temporarily increasing the workforce • Advancing or postponing individual operations
• Create anticipation inventory in periods of low demand • Reduce setup time
The creation of anticipation inventory of parts which are the same in each belt or baler seems less appropriate. For balers it is estimated that the total weekly savings are small, but the cost of waste can become relatively high. In case of sliding belts it can only be applied to single orders. It does not make sense to use this option for large projects, since demand is known far beforehand, and therefore will only lead to additional carrying costs. Since the demand for single order is so small, but still shows large variety in the widths and lengths it is a difficult decision on what to produce to stock. The risk of long storage duration is therefore again present.
The last option to reduce the setup time also seems an appropriate solution. By hiring temporary employees which can aid in removing the parts and plates from the machine the setup time can possibly be reduced by 50%, which removes an additional 60,5 hours of load. The utilization level drops to 73,87%, the idle capacity is in this case 334,45 hours. This idle capacity can in turn be used to also insource the frames of chain and trough belts. However even if the setup is not reduced there still is 273,95 hours left to cut these types of belts.
Concluding it can be said that it is certainly an option to insource the cutting of the frames of sliding belts, since balers only require 30/35% of the total capacity at cutting. It makes BRM certainly more flexible since they now can decide when to start each order and are not dependent on the uncertainty at the supplier side anymore. Besides that there is plenty of capacity left to insource at least a part of the trough and chain belts.
It is however important that capacity is monitored carefully, if no actions are taken substantial overloads can occur. The best option seems to reschedule those orders which are part of a large project, since these are known far beforehand. In order to avoid the carrying cost from becoming too high the sliding belts should only be brought forward a couple of weeks.
III TABLE OF CONTENTS
I. Preface 2
II. Management summary 3
III.Table of contents 7 1. INTRODUCTION 9 1.1 Company description 9 1.2 Problem exploration 11 1.3 Research objective 17 1.4 Research question 18 1.5 Conceptual model 19
1.6 Data collection methods 20
2. PRODUCTION PLANNING AND CONTROL AT BRM 22
2.1 Performance objectives 22
2.2 Planning and control characteristics 27
2.2.1 Features pertaining to the user and product or
product family 27
2.2.2 Features in reference to logistics and production
resources 30
2.2.3 Features in reference to production or
procurement orders 32
2.3 Production planning and control system at BRM 33
2.4 The applicability of MRP for BRM 34
2.5 Conclusion 37
3. THE PRODUCTION PROCESS 38
3.1 Current production process balers 38
3.2 Manufacturing lead time 41
3.3 Changes in the production process 45
3.4 Conclusion 48
4. THE LONG TERM PLANNING 49
4.1 Demand 49
4.4 Master production schedule 54
4.5 Conclusion 55
5. THE OPTIMAL BATCH SIZE 57
5.1 Important considerations with regard to the batch size 57 5.2 Comparison of the three batch size options 61
5.3 Conclusion 67
6. THE INFLUENCE OF INSOURCING SLIDING BELTS 68
6.1 Production process sliding belts 68
6.2 MPS sliding belts 70
6.3 Rough cut capacity planning 71
6.3.1 Rough cut capacity planning methods 71
6.3.2 Rough cut capacity planning cutting 73
6.3.3 Actions to avoid overload 76
6.4 Conclusion 82
7. CONCLUSION AND RECOMMENDATIONS 85
7.1 Conclusion 85
7.2 Recommendations 88
7.3 Reflection 89
References 90
1. INTRODUCTION
This part of the report starts with a company description, including the products Bollegraaf Recycling Machinery offers. Next the problem statement, research goals, research question and finally the research methodology will be described.
1.1 Company description
Bollegraaf Recycling Machinery (BRM) is situated in Appingedam, currently there are approximately 200 people working at BRM. The company was founded in 1969, ever since they have grown to one of the leading companies in the recycling machinery industry. A “Bollegraaf” is considered to be the Rolls Royce of the recycling machinery. BRM supplies their recycling machinery worldwide to the waste paper industry, to collectors of household and industrial waste, to municipalities and to companies which produce voluminous waste. Both machines as entire waste system solutions are delivered to their customers worldwide. BRM has two subsidiaries in the Netherlands, located in Emmen and Rotterdam, and four foreign subsidiaries in Germany, Spain, France and England. Besides that they are participating in dealer organizations in the USA and Canada. These subsidiaries are all on behalf of the sales and services of the recycling machines.
Products
The machines which are included in BRM’s product range are: balers, sorting systems, conveyors, shredders, starscreens, reel splitters and other related equipment. As stated before BRM delivers both single machines as well as entire solutions for their customers. These solutions generally consist of a baler, sorting systems, and a number of conveyor belts to transport the waste.
BRM produces mainly balers and conveyor belts. There are three series of balers, namely: the HBC, HBK, and SA series.
• The HBC balers are used for wastepaper, cardboard, synthetics and cans. The capacity and compaction force of the HBC balers ranges from 25 to 180 tonnes.
balers have a large capacity but do not need much space for their installation. A conveyor belt can be linked in such a way that material can be put on the belt from all sides.
• The SA balers are made for compacting all types of bulky waste such as waste paper, plastics, textiles, rubber and leather. It can easily deal with this range of materials without employing a shredder. The filler opening is at hand height, making processing bulky materials very simple.
All these types of balers can be equipped with a large number of options like a bottle perforator, turbo press and weighting unit. Beside all these aspects do the machines also vary in the amount of compaction force.
The conveyor belt product range consists of 4 types of belts: sliding belts, chain belts, steel belts and trough belts.
• Sliding belts are used to carry light, but large-size materials. They have a synthetic layer on the underside and the belt slides through a steel groove. The belt is put into motion by way of a drum driven by a motor. Sliding belts can be supplied in 600 mm to 2800 mm widths.
• Chain belt conveyors are used to carry waste paper, domestic and industrial waste. Chain belts use so called ‘grippers’ to transport the waste to higher levels in the waste recycling process. They can vary in regard to length, speed, heaviness and widths.
• Steel belts are designed for companies who need to transport heavy material of domestic of industrial waste. Due to a heavy steel construction and the special shape of the belt it can accept this heavy load. Again they can vary in length, speed, heaviness and widths.
1.2 Problem exploration
The recycling machinery industry has a turbulent environment. Demand is lumpy, competition is hardening and customers are putting more and more pressure on lead times. Another important change in the market is that municipalities want to stay within their budget, the lower they can keep the cost the better, and therefore pressure on prices is also increasing. In order to respond to these changing conditions, BRM has decided to invest in two new machines. The new equipment consists of an automatic welding line and two plasma cutting machines which will be used to construct the basic framework of balers. Besides that are the new cutting machines also able to cut sheet metal for the different conveyor belts. The plasma cutting machine will replace the old cutting machine. The two new machines both have two burners for cutting, while the old autogenous cutting machines had four burners which were dependent on each other. Thus the capacity can be used more flexible and it enables BRM to cut in smaller batches. The automatic welding line will partially replace manual welding, some parts will still require manual welding since the machine is not able to weld every specific part. The cutting machines are used at the first step of the production process of balers, where the different metal parts are cut in the appropriate size for each type of baler. At welding the different parts are welded together to create the framework of a baler. This is now done manually, but will be done mostly automatically by the new welding line. This research will focus on the two cutting machines.
Obviously these new machines have more efficient production techniques than the older machines, leading to shorter operation times. Hence overcapacity arises when the normal logistical procedures remain. BRM wants to know how to utilize this capacity most optimally. Capacity utilization is a measure on how intensively a resource is being used to produce a good or service. Traditionally it is the ratio of actual load to theoretical capacity (Schönsleben, 2007).
temporal ranges are often displayed using the MRPII/ERP framework (Schönsleben, 2007). The planning hierarchy (figure 1.1) in MRPII consists of three planning dimensions: the long-term, medium-term and short term planning. These are also known as front-end, engine, and back end (Vollmann, Berry, Whybark, & Jacobs, 2005).
FIGURE 1.1
Production planning and control framework
Long-term planning and control
Long-term planning, also known as master planning, aims to forecast the total demand for products and processes that will be placed on the enterprise or logistic network (Schönsleben, 2007). Based on this forecast the company can derive quantities and gain the resources necessary to fulfil demand. The planning horizon usually expands from several months to a year. The decision to purchase the new machines is part of the long-term capacity planning.
Demand management
Sales & operations planning Master Production Scheduling Resource requirements planning Rough-cut capacity planning Detailed planning and scheduling Detailed capacity planning
Shop floor control Supplier systems Long-term planning
Medium-term planning
The first task in long term planning is that of demand management. Demand management is the function of recognizing all demands for goods and services to support the marketplace (Schönsleben, 2007).
Based on this demand forecast, the sales and operations planning (S&OP) is formulated. The operations and sales plan links all the tactical plans for the business with it executions. In early MRPII terminology the term production planning was used instead of S&OP (Olhager, Rudberg, & Wikner, 2001). The most important processes with regard to operations management are the sales plan, production plan, stock inventory plan, and the procurement plan.
The sales plan represents the expected customer orders to be received for each major product family. It is management’s commitment to achieve this level of customer orders, can be dependent on the forecast and is expressed in units on an aggregate level (Schönsleben, 2007). Based on this sales plan the production plan, procurement plan, and inventory plan are developed. A production plan reflects how a company tries to respond to the various demand levels of the sales plan. If the demand pattern cannot be changed, there are three planning strategies a company can follow: a level plan, a chase plan, or a mix plan (Olhager et al., 2001). A level plan ignores the demand fluctuations and keeps activity levels constant (Slack, Chambers, & Johnston, 2004). A chase plan on the other hand implies that production matches demand in such a manner that all demand in a period is produced in the same period. This does require quantitative flexibility of capacity. In the mix plan a production rate is used for a few periods and then changed.
Any difference between the production plan and sales plan results in an inventory or backlog plan (Olhager et al., 2001). In case of a make-to-stock (MTS) environment this will be an inventory plan, in a make-to-order (MTO) or engineer-to-order (ETO) environment this is a backlog plan. The inventory plan determines the desired level of stored items, according to the company’s inventory policy.
requirements (procurement plan) and capacity requirements, and the cost budgets (Schönsleben, 2007).
The sales and operation planning mainly works with aggregate information, however there is also a need for more specific information, this leads to the third task, which is master scheduling and rough-cut capacity planning. Master scheduling is the planning process at individual level, and results in the master production schedule (MPS), which is the disaggregate version of the production plan. The MPS does not only show the individual products instead of product families, but also more detail at which time the products are aggregated. The MPS is therefore seen as the link between the production plan and the products manufacturing will actually build, it is the input to all planning actions in shorter term (Schönsleben, 2007). One way to verify the feasibility of the MPS is by rough-cut capacity planning (RCCP). At RCCP the MPS is converted into required capacity, that is capacity of key resources to produce the desired output in the particular periods (Schönsleben, 2007). RCCP yields more precise information on the work centers capacities to be used than does resource requirement planning. It shows the load of the MPS in comparison with the weekly capacity of each work center, and thus the utilization levels. There are three different RCCP techniques at the long term level: capacity planning using overall factors (CPOF), capacity bills, and resource profiles (Jonsson & Mattsson, 2003; Berry, Schmitt, & Vollmann, 1982). These vary in accuracy, where resource profiles generates the best results, it also requires the most data (Hendry & Kingsman, 1989).
Medium-term planning and control
Materials management aims at providing the goods required by demand both cost effectively and according to schedule. Some of its main objectives are to avoid disruptions in delivery or production, and to keep the carrying costs as low as possible. Time management and scheduling, and capacity management aim at determining which capacities need to be available, how individual order processing tasks can be synchronized, and where and when special shifts and overtime must be put in place (Schönsleben, 2007).
Short-term planning and control
The last dimension is short-term planning which deals with the execution and control of operations. Here control takes the form of coordination, which is performed by all persons involved (Schönsleben, 2007). Most of the short-term decisions take place at the shop-floor, and considers for example the exact timing of machine changeovers and the actual assignment of orders to machines (Van Wezel, Van Donk, & Gaalman, 2006).
With the regard to the scheduling of orders there are three methods commonly used, infinite scheduling, finite scheduling, and input/output control (Jonsson & Mattsson, 2003). The first does not consider shop load when releasing the orders to the floor and is therefore the simplest. The second does consider shop load and tries to avoid over- or underload. In situations with long lead times, many operations, and a functional layout it becomes difficult to avoid over- or underload, in this case input/output control is a good method.
1.3 Research Objective
The new machines were acquired because of the changing market conditions. Demand is hard to predict, and customers are putting more and more pressure on lead times and costs. The key purpose of the production planning and control functions is to reduce work-in-process (WIP), minimize shop floor throughput times and lead times, lower stockholding cost, improve responsiveness to changes in demand and improve delivery date adherence (Stevenson, Hendry, & Kingsman, 2005).
An important determinant of these objectives is the lot size decision. The lot size decision determines the amount of setup time (and so the capacity utilization for given production quantities) and the arrival rates of orders at capacity units. Therefore lot-sizing influences the amount of WIP necessary to realize certain production quantities. This WIP in turn determines the average flow times, since more WIP means longer waiting queues, which in turn influences the due date performance (Zäpfel & Missbauer, 1993). Thus we need to consider the influence of the batch size on the WIP level. In contrast with the autogenous cutting machine, which was only able to cut in batches of four, the two new machines are able to cut the parts of balers in batches of one or two. Therefore the main objective is to determine which batch size satisfies demand with regard to the performance objectives most optimally.
As described above the new cutting machines are able to cut different parts of conveyor belts as well, BRM also wants to know what the influence of this additional load on the cutting machines will be.
Based on the outcomes they can decide whether or not it is feasible to also insource the other types of conveyor belts.
1.4 Research Question
Based on the problem exploration and research objective the following research question is formulated:
“What is the optimal batch size of the new cutting machines with regard to balers and what will be the influence of the additional cutting for sliding belts?”
In order to give an answer to this question a number of sub-questions need to be answered, which are derived from the problem exploration. The first aspect that needs to be considered are the planning and control characteristics of BRM and the most important performance objectives. These characteristics influence the long and medium term planning, and also help to determine the appropriate capacity management technique. Therefore the following questions are formulated:
1. What are the most important performance objectives? 2. How is production planned and controlled at BRM?
The second part consists of determining the change in the production process caused by the new machines. Next to that does the production process give an insight into how the lead times of balers are built up. This results in the following questions:
3. Which changes occur in the production process?
As soon as both the planning and control characteristics, and production process are described the long term planning can be made, which will be based on historical data. The long term planning will provide the starting data for the different scenarios. As described is the main goal of long term planning to determine which demand is placed on the enterprise and how this demand can best be fulfilled. Therefore the following questions need to be answered:
4. Which demand is placed on BRM and how does BRM try to fulfil this demand?
sliding belts that need to be cut. First the optimal batch size of balers will be determined.
5. What is the optimal batch size for balers with regard to the most important performance objectives?
The most appropriate batch size will be the one which performs best at both performance objectives. The last step is to determine what the influence of the additional cutting of sliding belts will be. For this a rough cut capacity plan will be created to show the utilization level and loading pattern. This pattern will then be analyzed. This results in the last sub question:
6. What is the influence of the additional cutting of sliding belts?
1.5 Conceptual model
FIGURE 1.2 Conceptual model
Demand Forecast
Production planning and control
Sales and operations planning Planning and control characteristics Master production schedule (MPS) Batch size calculation balers Influence cutting for sliding belts
Long term planning
Optimal batch size (speed/cost)
The above displayed conceptual model shows in which way the research will be performed. As can be seen it is based on the production planning and control framework of figure 1.1 and is part of the long term planning. Before the main- and sub questions can be answered first necessary data must be collected. Most of this data will be historical data. The data collection methods are described in the next part of the chapter.
1.6 Data collection methods
All data collection methods are summarized in table 1.1
TABLE 1.1
Data collection methods
Required data Collection method Place of collecting data
Planning and control at BRM Literature study, Interviews
Articles and books, planners, &
logistical manager. Production process Data analysis,
Interviews
Production manager, logistical manager, & production department. Long term planning Literature study,
Data analysis, Interviews
Articles and books, planners, sales, & logistical manager.
Batch sizing Literature study
Data analysis, Interviews
Articles and books, production manager, & logistical manager, Rough cut capacity planning Literature study
Data analysis, Interviews
Articles and books, logistical manager, & planners
Literature study
Interviews
The interviews will be used to get a better understanding of the organization. The production manager will be interviewed to get an understanding of the production process. The logistical manager and planners will be interviewed to gain more insight on production planning and control at BRM, as well as to gain the required data.
Data analysis
2. PRODUCTION PLANNING AND CONTROL AT BRM
This chapter deals with how capacity is planned and controlled at BRM. The values of the typical characteristics for planning and control will vary depending on the performance objectives on which the company emphasizes. From these values the appropriate techniques from the two classes of finite and infinite loading can be derived (Schönsleben, 2007). First the main performance objectives of the organization will be discussed, followed by the different planning and control characteristics as described by Schönsleben (2007). Finally the production planning and control system of BRM will be described.
2.1 Performance objectives
Slack and Lewis (2002) consider five performance objectives important for an operations strategy. These performance objectives satisfy market requirements, and obviously differ in their relative priority for each organization. The following five objectives are defined: cost, quality, speed, dependability, and flexibility. All five objectives will be explained more in depth and related to BRM, as well as to the role which the new machines can play in achieving these objectives.
Cost
For companies competing directly on price, cost will be their main performance objective. However for all other companies it is also interesting to keep their costs as low as possible. Every euro saved at operations is an extra euro of profit. For capacity planning and control, costs will be affected by the balance between capacity and demand. If a company keeps higher capacity levels than the current market demand, underutilization occurs leading to higher unit costs. Working capital will be affected if an operation decides to build up finished goods inventory prior to demand (Slack et al. 2004).
Quality
There are many ways to define quality, Slack and Lewis (2002) divide quality in specification quality and conformance quality. Specification quality is on turn divided into hard and soft dimensions. Hard dimensions are the objective dimensions of the product, like reliability, performance, its features etc. The soft dimensions relate to the personal feelings one has with the products, aspects like friendliness, helpfulness and attentiveness are considered. Conformance quality is more related to operations, it refers to the ability of an operation to produce goods according to their defined specification reliably and consistently. Quality can be influenced if an organization experiences large fluctuations in capacity levels. It might be necessary to hire temporary staff. The new staff and disruption in routine work might lead to an increase in errors (Slack et al. 2004).
As stated earlier is a ‘Bollegraaf’ considered to be the Rolls Royce of the recycling industry. This is because of the high quality delivered, but also because of the somewhat higher price then the competitors. The aspects on which BRM distinguishes themselves from their competitors is the number of employees needed to operate the machine and the low maintenance cost. The initial price might be higher, but there are only a couple employees needed to operate the machine, and due to the high quality, maintenance costs are lowered. Therefore quality is the order winner for BRM.
It is expected that the new machines aid in achieving a higher quality level. Due to more efficient techniques less deburring and grinding is required anymore and the waste is also minimized. Thus less manual work is required, and therefore a decrease in errors is expected.
Dependability
Dependability is an important performance objective, but it is not an order winner for BRM. The lead times are long in the recycling machinery industry. When an order is placed it takes around 3 months for a baler to be delivered. For an entire waste system solution, lead times over half a year are normal, taking into consideration that the building where this system is placed also needs to be finished. However, customers want their products on time because of the high investment associated, but also because they will not be able to process their waste sufficiently if the machines are not on time. Therefore penalty costs are documented in contracts, not meeting due dates leads to high penalty costs. Being dependable is therefore an order qualifier, a company operating in the recycling machinery industry just needs to deliver their products on time, or else customers will be lost.
There is sufficient capacity, however due to inefficiencies dependability is lowering. Furthermore is BRM very dependable on on-time delivery of their suppliers, since this currently is not the case this also influences dependability negatively. The new machines can have a positive effect on the dependability since shorter operation times might lead to shorter lead times. However, it needs to be considered that the additional load caused by the cutting for sliding belts can lead to overload which in turn might negatively influence the due date. It is important that there is sufficient capacity to complete all the jobs at their promised due date. Thus when determining the optimal utilization level of the cutting machine, it is important that dependability is not lowered.
Speed
Speed indicates the time between the beginning of an operations process and its end (Slack & Lewis, 2002). This may relate to the time between an order entering the factory till delivery to the customer. But it can also be defined as the time between materials entering the operation till leaving fully processed.
Flexibility
The last performance dimension is flexibility. Slack and Lewis (2002) define two dimensions for four types of flexibility. The two dimensions are range and response flexibility. Range flexibility refers to how much the operation can be changed, response flexibility refers to how fast the operation can be changed. The four types of flexibility they distinguish are:
• product or service flexibility (the ability to introduce and produce new products or modify existing ones),
• mix flexibility (the ability to change the variety of products) • volume flexibility (the ability to change the level of output)
• delivery flexibility (the ability to change planned or assumed delivery dates). Especially volume flexibility will be enhanced by keeping excess capacity. If demand and capacity are in balance, the operation will not be able to respond to any unexpected changes in demand (Slack et al. 2004).
When looking at the four types of flexibility, volume flexibility is deemed to be the most important. For BRM it is hard to change the output. When demand and capacity are in balance an operation will not be able to respond to any changes in demand. Since demand is changing a lot in the recycling industry BRM needs to be flexible.
If a company cannot respond to the different customer requests they will lose sales, therefore flexibility is an order winner.
Especially the new cutting machine makes BRM more flexible. First of all BRM now has two cutting machines instead of one, which enable BRM to produce in smaller batches. Next to that they are now also able to produce parts for conveyor belts, which would normally have been outsourced.
Below a polar representation is made of the relative importance of each performance objective.
FIGURE 2.1
Performance dimensions BRM
The green line shows the current state of performance objectives. The yellow line shows the expected change in performance when the new machines are in use. As can be seen it is expected that especially speed will be increased. Partially because there is more flexibility due to a higher capacity. Due to the fact that the manual work is replaced by automated work there is also an increase in quality expected. Cost might be lowered due to lower capital costs, however the depreciation costs might negate this effect, therefore no change in the total cost level is expected. It is also not expected that insourcing is cheaper than outsourcing sliding belts frames.
2.2 Planning and control characteristics
Schönsleben (2007) provides an extensive list of characteristics of planning and control. These will be applied to the situation of BRM. This list is described for the product family balers, since the new machines will certainly be used to produce balers. The production process itself will be described in the next part.
Schönsleben (2007) divides the features into three groups: features pertaining to the user and product or product family, features pertaining to logistics and production resources, and features pertaining to the production or procurement orders. All these features will be described more in depth below.
2.2.1 Features pertaining to the user and product or product family
Depth of the product structure
The depth of the product structure is defined as the number of structure levels within the total logistics network for the product. Balers consist of four structure levels, these consist of seven main parts which in turn consist of a different number of parts. The seven main parts are:
• Frame: 156 parts • Pinch-off: 70 parts, • Stamp: 82 parts • Front-valve: 79 parts • Protection: 110 parts • Wiring-unit: 70 parts • Hydro-unit: 77 parts
Orientation of the product structure
This relates to the fact that one single product is manufactured from various components or that various products are made out of a certain component. Balers have a convergent production structure. As described above does a baler consist of seven main parts which are assembled together. These parts are the framework, pinch-off, stamp, front-valve, parts for protection, wiring and hydraulics. They are assembled together at end assembly. The needle and knotting installation are also added here.
Frequency of customer demand
The frequency of customer demand refers to the number of times within a defined observation time period, that the entirety of the customers demand a product or product family. There is no certain pattern in demand, the frequency in customer demand can therefore be classified as lumpy. In one month the demand can be 5 balers while in the other month no balers are ordered at all. On average 3,5 balers are delivered each month, based on the figures of 2007 and 2008. The demand is analyzed more in depth in the next chapter.
Product variety concept
The product variety concept determines the strategy for developing the product and offering it to the customer. Balers are defined as a product family, there are a number of different types of balers, varying in compaction force, which can be equipped with or without cross-wiring. The compaction force is related to the balers name, e.g. an HBC120 has a compaction force of 120 tonne. Next to that there are around 20 different options a customer has which can be added to the baler.
Unit costs
A baler is a high cost item, table 2.1 shows the different prices of each baler, taken from the ERP system Axapata.
The cost price is divided into three parts: the cost to construct the basic framework, the cost of assembly, and the costs of the unit used to power the baler. The first two are a combination of materials costs and the number of hours needed to produce the baler.
TABLE 2.1
Cost prices balers according to Axapta
Framework End-assembly Unit Total costs
HBC 60 x x x x HBC 60K x x x x HBC 80 x x x x HBC 80K x x x x HBC 100 x x x x HBC 100K x x x x HBC 120 x x x x HBC 120K x x x x HBC 110 x x x x HBC 110K x x x x HBC 140 x x x x HBC 140K x x x x HBC 140M x x x x HBC 140MK x x x x HBC 180 x x x x HBC 180K x x x x HBC 180M x x x x HBC 180MK x x x x
Each operation has a specific hourly cost rate, thus the labor costs of each part are calculated by multiplying the hourly rate of each operation with the hours required to perform this operation. These are the same as the hours specified at the lead times of each baler.
While the labor costs do not increase much from the cheapest baler to the most expensive, the material costs do increase rapidly. E.g. the labor costs for the HBC60 are approximately € X against € X for the HBC180. The material costs however are around € X for the HBC60, but almost € X euro for the HBC180, not including the costs of the engine.
to be. The HBC140 and HBC180 both require two engines, this is also one of the reasons why these balers are a lot more expensive.
Transportability
The transportability of an item is a statement on the size and weight per unit of measurement. A finished HBC180K weighs 49.600 kg. Looking at the definitions of Schönsleben (2007) balers can be classified as transportable items. However, technical aids are needed. The factory is equipped with heavy cranes to transport the parts. Besides that are forklift trucks with trailers needed to transport the machines between the two locations of BRM and Smit Coating, where they are sprayed and blasted. Based on the delivery destination the balers are either transported by truck (Europe) or by ship (USA). A baler is large to such an extent that only one baler fits on a truck or in a sea container.
2.2.2 Features in reference to logistics and production resources
Order penetration point
BRM only produces products when they are demanded by customers, therefore BRM can be classified as a Make-to-Order (MTO) organization. Based on the specific needs of the customer the order is started, therefore the order penetration point (OPP) lies at the first operation: cutting. A customer can for example choose between a baler with or without cross-wiring. Besides that can a different number of options be added to the baler at the assembly stage as well as the desired color. However due to the fact that the current cutting machine has four autogenous burners there automatically will be four balers in production. Thus three of them will be placed in stock before the final assembly, where the specific options are added. Thus when a customer orders a baler for which the framework is already in stock the production environment can also be Assemble-to-Order (ATO).
Depth of product structure in the company
The depth of the product structure in the company is defined as the number of structure levels within the company. The greater the number of structure levels the company itself produces (make decisions), the fewer components will be purchased from outside suppliers, and the greater the depth of the product structure in the company. Therefore BRM can be classified as a company with a deep structure. As stated above do balers have a deep product structure, most of its components are produced internally. The other products BRM delivers also have deep product structures. To reduce complexity certain production steps are outsourced, for balers this is the coating process.
Facility layout
The layout describes the physical organization of the production infrastructure, the degree of division of labor, and the course that orders take through the work centers. When looking at the different process layouts described the layout at BRM can be classified as a process layout. In a process layout similar processes are located together. The products flow through the production process from activity to activity, based on their needs. This is what happens at BRM. The balers follow a specific route according to the operations that need to be performed. The final step in the production process, end-assembly, can be classified as a fixed position layout. A pre-assembled baler is to such a degree heavy and large that instead of moving the product to each operation, the employees move around the product with toolboxes.
Qualitative flexibility of capacity
Quantitative flexibility of capacity
The quantitative flexibility of capacity describes its temporal flexibility along the time axis. It is crucial in choosing a planning and control method, particularly for capacity management. As described earlier, employees can be moved from one work center to another, depending on the load. Therefore the quantitative flexibility of employees is high. Quantitative flexibility of machines can only be achieved by maintaining over-capacity. BRM wants to move to the state where quantitative flexibility of machines is also high.
The only processes that do not have flexible capacity are pre- and end-assembly. Here a lot of specific knowledge is required and therefore it is hard to add additional employees at this process.
2.2.3 Features in reference to the production or procurement orders
Reasons for order release
The reason for order release is the origin of demand. In the case of BRM order release is according to demand. Orders are released according to delivery agreements. Thus the production is pull-controlled.
Frequency of order repetition
As the name implies, this deals with how often a production or procurement order for the same product will be made within a certain time period. There is a certain probability that an order for a specific type of baler will be placed again. However it cannot be said that an order will be placed very frequent due to the lumpy demand characteristics. Hence the order repetition can be called infrequent.
Flexibility of the order due date
Type of long-term order
BRM uses long-term orders for the supply of raw materials. At the beginning of the year they make a forecast on the number of balers expected to be sold. Based on this forecast they reserve an amount of sheet metal at the supplier. Furthermore it might happen that a customer orders an entire waste system, consisting of multiple balers, conveyor belts, starscreens etc. These projects usually take more than half year, and guarantee a certain demand for this period, since sales are assured.
Lot size or batch size of the order
The lot size or batch size of the order is the order quantity of an ordered item. For BRM this depends on the type of order. Normally this is single-item production, only one baler is produced for an order, however as described earlier does BRM produce in batches of four. Thus when an order occurs, one baler will be completed and the parts of three balers will be placed in stock. In the case of a waste system solution more balers might be required. The new machines are able to cut in smaller batches, leading to a batch size of one or two.
2.3 Production planning and control method at BRM
At BRM production is planned and controlled using the ERP package Axapta. The production planning and control hierarchy is the same as in figure 1.1, displayed at the problem exploration. At the higher level orders are accepted and transferred to a MPS. This MPS is created using rule of thumbs, ensuring that available capacity is not exceeded. For BRM this means that they are not able to deliver more than one baler each week, and a maximum of 10 conveyor belts.
At the medium term level MRP is used to calculate the different material and capacity requirements. MRP calculates the dependent demand by exploding the bill of material of the higher level independent demand (Schönsleben, 2007).
at which it will also be released to the shop floor. At this stage the drawing made by engineering needs to be ready.
After scheduling all the orders, each scheduled operation represents a load at the specified work centers and its start date. The sum off all the scheduled loads is compared to the available capacity for each period. In case of an exceptional situation the planner can decide to bring the earliest start date forward, if possible. Based on the resulting schedule production orders are generated by the work preparation department. Each day a list with the specific tasks that need to be performed is generated by Axapta. This list is sent to the foremen who will decide how to utilize his employees. If it for example happens that almost no work is required at deburring and grinding, but a lot at drilling and punching he will move employees between these two operations. The sequencing at shop floor level is therefore also performed by the foremen, who have a good grasp on the actual workloads
Besides that is for each part of the bill-of-material an availability check performed on daily basis. If certain parts are not on hand the purchasers will generate purchasing orders for these parts. Eight weeks before the start of the production the first materials need to be purchased due to the long lead time offset of certain parts, like the engine. This generation of production and purchasing orders is part of planning and control at the short term level.
2.4 Conclusion
We can now give an answer to the first two sub questions “what are the most important performance objectives?” and “how is production planned and controlled at BRM?”.
3. THE PRODUCTION PROCESS
Now that the planning and control characteristics of BRM are described, it is time to describe the changes which occur in the production process. First the production process of balers in the current situation will be described. Next the manufacturing lead times will be described, again as how they are in the situation without the new machines. This chapter will end with the describing the changes which occur in the production process.
3.1 The current production process of balers.
The current production process of balers is displayed in figure 3.11. As described earlier balers are convergent products. This means that multiple parts are assembled together into one final product. A baler consists of the following seven main parts:
• Frame • Front-valve • Stamp • Pinch-off • Hydraulics • Protection parts • Wiring
The first four parts are all constructed at production site called the ‘stadsweg’ and follow almost identical routes through the production process. There are some small differences in the processes, e.g. a pinch-off does not require any shearing or mounting. However, this does not influence the entire throughput time, since these four parts are produced parallel to each other. Therefore the critical path determines the entire throughput time. The stamp has the longest manufacturing lead time and is therefore part of the critical path. The order penetration point can be at two places. When the ordered baler is not in stock the order penetration point lies at the beginning of the production process. In case there are frameworks of this baler in
1
stock the order penetration point lies behind welding. At welding the front-valve and pinch-off are welded to the frame. When this is completed, the frame and stamp will be transported to Smit Coating together with the stamp, where they will be sprayed and blasted. When this is finished the baler is transported to the ‘tweede industrieweg’, here the last three parts are added at pre- and end- assembly. The different options the customer requires are also added. When the pre-assembly is finished the baler will be painted in the desired color. Below the different production steps are described more in detail.
Cutting
The first step in the production process is cutting. A wagon containing a steel plate is towed into the cutting machine where it is cut into the appropriate size. In the current situation there are two cutting tables and four autogenous burners. These burners are dependent on each other, therefore BRM produces in batches of four. One on customer order, the other three on stock. When the cutting process is finished the plates are towed into the main production hall where the plates will be deburred. To reduce waste the parts of the frame, front-valve, stamp and pinch-off are cut from the same steel plates, in batches of four.
Deburring & Grinding
After the sheets have been cut the different parts need to be deburred. At deburring the sharp edges are removed from the steel parts. This is all done manually using a portable grinding machine. When the deburring is finished the parts are put into storage outside.
Sawing
Drilling & Punching
When the parts are deburred and possibly cut into the right shape they are transferred to drilling and punching. Here, as the name implies, holes are drilled into the steel plates.
Cottering
Cottering is not part of the production process of frames (see footnote 1). However certain parts of the front valve and stamp need cottering. It is performed at the ‘Tweede Industrieweg’.
Milling & Drifting
At milling and drifting the hinges of balers are chipped up into the right size. There is one lathe used for drifting and one milling cutter.
Shearing & Mounting
Shearing and mounting has its own automated storage where a number of plates are stored. The storage delivers the required plates which are sheared after they have been marked out. Depending on what is required next, the plates are punched or notched followed by the mounting process.
Note: The shearing and mounting is performed at the “Tweede Industrieweg”, however due to the fact that it is part of the production process of the first four parts, it is for clarifying reasons displayed at the “Stadsweg” in figure 4.
Construction benchworking heavy (CBW)
At construction benchworking ‘heavy’ the frame, front-valve, pinch-off and stamp are constructed. Each part is constructed at a specific workshop. Constructing the frame is most time consuming, therefore a maximum of five employees are assigned here. At the construction of the other three parts only one employee is required for each part.
Welding
be transported to Smit Coating together with the stamp and a box containing hydraulic and protection parts. Here it will be sprayed and blasted, which will protect the baler against corrosion and pollution.
Pre-Assembly
The sprayed and blasted framework and stamp are transported to pre-assembly. Pre-Assembly is a complex process where a lot of different operations are performed and therefore specific knowledge is required. The first part consist of tagging the machine and preparing it for painting. Next to that are some parts of hydraulics, protection and wiring added to the baler. The stamp is also mounted onto the baler here. When this is finished the pre-assembled baler will be transported to painting.
Painting
As the name implies the products are spray painted here. Normally BRM uses their “home-colors” green/yellow when painting a baler. However the product can be painted in every color the customer desires. The small parts are sprayed while hanging on a rail, the bigger parts are spray painted while lying on a trailer. The painting is al done manually. The sprayed parts are mostly dried outside since BRM does not have a dry-installation.
End-assembly
The end-assembly is the final part of the process. At end-assembly especially a lot of protection parts are added like the protection cage and protective covers. Besides that are the needle- and knotting-installation added here. The end-assembly is performed in the main production hall of BRM. The standard is that one machine is finished each week here.
3.2 Manufacturing lead time
According to Schönsleben (2007) the manufacturing lead time consists of operation time, interoperation time and administration time. The operation time consists of the time required to carry out a particular operation. It consists of the setup time and the run time for the actual order lot. The setup time is independent of lot size, the run time is dependent on the actual lot. The interoperation time occurs before or after an operation. Interoperation time can consists of a different number of factors. It can consist for example of technical wait time after an operation, e.g. testing. Besides that it can consist of transportation time or non technical wait time due to queues. The operation times (in hours) of the different types of balers without cross-wiring are displayed in table 3.1. Table 3.2 shows the operation times of balers with cross-wiring.
In these tables the times to construct one baler are taken, the batch size is therefore considered to be one, also for cutting and sawing. The reason for this is that this is the actual time needed to construct one baler. If the batch size of four would be taken, the lead time of the first baler of the batch would be longer, while the lead times of the second, third, and fourth in the batch would be shorter.
TO stands for total operation time of one baler, this is the sum of the hours required for each operation and all the interoperation times. The critical path (CP) for each baler is displayed in bold. Slack (2004) defines the critical path as ‘the path that has the longest sequence of activities’. It is called the critical path because any delay in any of the activities on this path will delay the whole project. Looking at the parts produced at the “Stadsweg”, the framework has the longest lead time. Hence, the time needed to construct a finished framework is part of the critical path and therefore directly influences the total throughput time.
TABLE 3.1
Operation times balers without cross-wiring
Operation times balers without cross-wiring Fra m e F ro n t-v a lv e P in c h -o ff S ta m p V S m it C o a ti n g A s s e m b ly P a in ti n g L e a d t im e T o ta l O T HBC 60 TO 250.38 114.67 45.83 93.67 40 166.67 20 731.22 CP 222.97 40 166.67 20 449.64 HBC 80 TO 284.25 115.17 47.55 102.5 40 166.67 20 776.14 CP 255.51 40 166.67 20 482.18 HBC 100/120 TO 299.99 117.83 52.49 107.08 40 166.67 20 804.06 CP 270.83 40 166.67 20 497.5 HBC 110/120s TO 331.31 117.83 61.83 111.74 40 166.67 20 849.38 CP 302.15 40 166.67 20 528.82 HBC 140/180 TO 386.48 127.83 70.02 112.61 40 186.67 20 943.61 CP 355.4 40 186.67 20 602.07 HBC 140M/180M TO 383.83 127 70.17 113.96 40 186.67 20 941.63 CP 352.75 40 186.67 20 599.42 TABLE 3.2
Operation times balers with cross-wiring
Operation times balers with
cross-wiring Fra m e F ro n t-v a lv e P in c h -o ff S ta m p K S m it C o a ti n g A s s e m b ly P a in ti n g L e a d t im e T o ta l O T HBC 60K TO 250.88 114.67 45.83 100.42 40 200.67 22 774.47 CP 223.47 40 200.67 22 486.14 HBC 80K TO 285.25 115.17 47.55 116.67 40 200.67 22 827.31 CP 256.51 40 200.67 22 519.18 HBC 100/120K TO 300.33 117.83 52.49 117.42 40 200.67 22 850.74 CP 271.17 40 200.67 22 533.84 HBC 110K/s120sK TO 331.58 117.83 61.83 117.97 40 200.67 22 891.88 CP 302.42 40 200.67 22 565.09 HBC 140/180K TO 386.92 127.83 70.02 119.92 40 224.67 22 991.36 CP 355.84 40 224.67 22 642.51 HBC 140M/180MK TO 384.73 127 70.17 135.1 40 224.67 22 1003.7 CP 353.65 40 224.67 22 640.32
the time is even 21 hours longer. This difference is mostly caused by the operations CBW heavy and welding. The largest difference however is found at assembly, which takes around 34 till 38 hours longer. The reason for this is that a baler with cross-wiring requires an additional knotting- and needle-installation, and extra cross-wiring. Especially the addition of an extra needle-installation is time consuming, this takes around 24 hours.
The interoperation time is the same for each baler. BRM includes four hours of wait time after each operation, except for sawing, after which 12 hours of wait time is included. This time is added to cope with unforeseen disturbances, move the parts to the next work station, and to test certain parts. The reason that the time after sawing is longer is because of the disturbances that might occur. If the sheet metal is not of the right quality, it can happen that the sheet metal starts curling, if this occurs the plate needs to be straightened, which is time consuming.
When a baler is finished it is tested to determine if it working properly, this takes approximately four days. The total interoperation time for each baler is 160 hours, and is built up in the following way:
TABLE 3.3
Interoperation times balers
Interoperation times (in hours)
Frame 40 Front-valve 32 Pinch-off 16 Stamp V/K 28 Assembly 8 Painting 4 Testing 32 Total 160
TABLE 3.4 Operation times HBC80 C u tt in g S a w in g D e b u rr in g & G ri n d in g D ri lli n g & P u n c h in g S h e a ri n g & M o u n ti n g C o tt e ri n g M ill in g & D ri ft in g C B W H e a v y W e ld in g C o tt e ri n g C B W l ig h t S m it C o a ti n g P re -A s s e m b ly P a in ti n g E n d -A s s e m b ly Operation 200 100 210 120 150 160 170 320 330 160 300 400 460 550 Total Hours 24 12.24 28.84 33.5 0.83 1 8 193.17 122.89 5 4 40 46.67 16 80 616.14 CP 24 15.5 21.67 2 106.67 61.67 40 46.67 16 80 414.18 The operation times of the other balers are spread in the same ratio as the above displayed times.
In appendix 1, a detailed overview of the different operation times and interoperation times for each baler are displayed.
