WELL BEGUN IS HALF DONE: UNCERTAINTY DURING THE PRE-PRODUCTION STAGE IN
HIGH VOLUME PRODUCTION ETO
COMPANIES
Master thesis, Msc Supply Chain Management University of Groningen, Faculty of Economics and Business
June 23, 2014
Robin Tieberink Studentnumber: 2536870 e-mail: r.tieberink@student.rug.nl
Supervisor/university Dr. G. D. Soepenberg
Co-assessor/university Dr. J. Wijngaard
1
ACKNOWLEDGEMENTS
This research has been conducted at an ETO company in the eastern part of the Netherlands. I would like to thank the employees for their cooperation and contribution to this research. furthermore, I would like to thank my supervisor Dr. G. D. Soepenberg for his contribution and his helpful feedback during the research.
2
ABSTRACT
The goal of this paper is to provide insights in uncertainty during the pre-production stage of high volume production ETO companies. Little research on the pre-production stage in general has been conducted so far.
After a literature review which results in a conceptual model, an in depth single-case study has been done.
Semi-structured interviews showed that uncertainty of specifications is the most influential type of uncertainty in the pre-production stage of ETO companies. We further found that the cause for this can be found in the very first phases and comes to surface in the design and engineering phase. This means that companies should first focus on reducing uncertainty of product specifications during the customer enquiry phase when they want to reduce uncertainty in their pre-production stage. The scientific relevance is that this paper provides insights in the causes of several categories of uncertainty during the pre-production stage of ETO companies for further research on this topic.
3
TABLE OF CONTENTS
1. Introduction ... 4
2. Literature review ... 6
2.1 ETO companies ... 6
2.2 Pre-production stage ... 6
2.3 Uncertainty ... 8
2.4 Conceptual model ... 9
3. Methodology ... 11
3.1 Method selection ... 11
3.2 Case selection ... 11
3.3 Data collection ... 11
3.4 Data analysis ... 12
3.5 Research quality ... 12
4. Results ... 12
4.1 Description of the company ... 13
4.2 Required information per department ... 14
4.3 Interview results ... 15
5. Discussion ... 20
5.1 Customer enquiry phase... 20
5.2 Design and engineering phase ... 21
5.3 Job entry phase ... 22
6. Conclusion ... 23
7. Limitations and further research ... 24
Reference list ... 25
Appendices ... 27
Appendix A – Interview protocol ... 27
Appendix B – Management samenvatting ... 28
4
1. INTRODUCTION
Many engineer-to-order (ETO) companies experience problems regarding delivery reliability in the sense that lead times are not met (Pandit & Zhu, 2007). Problems can often be traced back to poor planning in the pre- production stage (Land & Gaalman, 2009). However, little research that is focused specifically on the pre- production stage has been conducted.
Bertrand and Muntslag (1993) presented a production control framework that suits the ETO environment. This paper is one of the few papers in which the pre-production stage is discussed. Most other papers on production planning and control in ETO or make-to-order (MTO) environments do not explicitly address the pre-production stage (Ebadian, Rabbani, Torabi, & Jolai, 2009; Henrich, Land, & Gaalman, 2004; Stevenson, Hendry, & Kingsman, 2005; Stevenson, Huang, Hendry, & Soepenberg, 2011). This can be seen as a significant gap in literature. This is emphasized by the findings of Land and Gaalman (2009) and Pandit and Zhu (2007) that the pre-production stage has great influence on delays and disruptions in the actual production process.
Land and Gaalman (2009) concluded that the problems regarding delivery reliability in ETO companies are often the result of planning problems in the pre-production stage. Two main problems can be distinguished in this stage according to Land & Gaalman (2009): (1) ‘Inadequate capacity planning overviews to support sales decisions’ and (2) uncontrolled delays in engineering. The master thesis from He (2013) addressed this problem and identified a few causes for delays in the pre-production stage. A major factor was found in the high level of uncertainty in this stage, this factor had most effect on the delays and even seemed to increase during the pre- production stage. Besides this, it causes a high level of complexity and dynamics in pre-production stage (He, 2013). Konijnendijk (1994) also found that companies in the ETO environment have to deal with high uncertainty and low controllability in their manufacturing processes. Uncertainty can thus be considered an important and influential factor in the pre-production stage of ETO companies, which makes it an interesting gap in literature to focus on in this study.
The aim of this research is to further investigate the uncertainty factor in the pre-production stage. The development and origin of uncertainty throughout the different phases of the pre-production stage will be investigated by means of the following research questions: (1) how can uncertainty in pre-production internally and externally be explained, and (2) during which activities in pre-production does uncertainty increase or decrease?
The research will focus on a specific type of ETO companies. Companies within this category may differ a lot (Hicks, McGovern & Earl, 2000). Earlier mentioned authors who wrote about pre-production like Land &
Gaalman (2009), Pandit & Zhu (2007) and Bertrand and Muntslag (1993) seem to focus on the classical job- shop type of company. However, this research will focus on companies with a high volume production stage including repeat orders after the pre-production stage, specifically in the metal forging industry. These high volumes of production make the pre-production stage even more important, as it is the basis for a product of which sometimes millions of pieces are manufactured.
An in-depth case study has been conducted to find the answers on the posed questions. The findings of this paper should help companies to diminish uncertainty in the pre-production stage. Besides this, the scientific relevance is that this paper provides insights in a main cause (uncertainty) of delays in the pre-production stage and help filling the literature gap mentioned by Land & Gaalman (2009). This provides more elaborated information on the uncertainty factor that is discussed by Bertrand and Muntslag (1993).
This thesis further contains a literature review in which the ETO environment and the pre-production stage in the ETO industry are discussed. Several types of uncertainty are considered and a conceptual model is presented to clarify the goal of the study. Then the methodology that is used to assure a thorough and reliable
5
research will follow, the results are presented and the findings are discussed in chapter five. The conclusion can be found in chapter six and lastly the limitations and opportunities for further research are discussed.
6
2. LITERATURE REVIEW
This literature review can be divided into four parts. The first part is a short description of ETO companies and their environments to specify what type of companies this research will consider. In the second part the pre- production stage is addressed and several stages within pre-production are described. This makes clear what the pre-production stage actually is and what activities are being done in this part of production. After this, several types of uncertainty that could be found in literature are described. Lastly, some relations between different types of uncertainty and the activities in pre-production stage are presented and a conceptual model is depicted.
2.1 ETO COMPANIES
Engineer-to-order companies may all fall in the same category, but the companies in this category might have completely different characteristics (Bertrand & Muntslag, 1993; Hicks et al., 2000). The type of companies this research focuses on, differs from the classical job-shop type of ETO companies. However, there are some more general characteristics or classifications that can be used to test whether a company can still be seen as an ETO company.
Gosling and Naim (2009) identified four unique characteristics of the ETO supply chain of which the first three are applicable in this research. They state that: (1) the production flow in an ETO supply chain is completely driven by actual customer orders, (2) the decoupling point is located in the design stage and (3) ETO companies can either develop completely new designs or adjust existing designs. The last characteristic is that several ways of organizing the ETO supply chain are present, but this would only be applicable if several companies where used in this research. Because only one company is used in this case study, we only have one way the supply chain has been organized.
Four different classifications for non-make-to-stock companies can be distinguished based on the customer specificity of the product (Amaro, Hendry, & Kingsman, 1999). These are: (1) pure customization, which means that the product is completely new developed and customized for the customer; (2) tailored customization, in which the customization encompasses a modification to an existing design; (3) standardized customization, which means that the company picks a design from an existing set of designs and (4) no customization/standard product, in which the design is taken and used as it is. Comparing these classifications with the third characteristic of Gosling and Naim (2009) that ETO companies can either develop completely new designs or adjust existing designs we can conclude that an ETO company in this study would fit in the pure customization or tailored customization category. Furthermore it should match the first two characteristics identified by Gosling and Naim (2009).
2.2 PRE-PRODUCTION STAGE
The pre-production stage of a company can be split up in different phases and activities. This division of activities in pre-production is important to assess the uncertainty during the flow through pre-production.
Literature is used to determine the phases in pre-production, then relevant activities within these stages are further defined.
A first distinction in the pre-production process in ETO companies can be made between a physical and a non- physical stage. The non-physical stage would include the activities that are finished early on in the production process: engineering, design and planning activities. The physical activities would include the actual production part: component manufacturing, assembly and installation of the machines (Bertrand & Muntslag, 1993). In this description, pre-production stage would include the non-physical activities.
A more precise distinction is made by Aslan, Stevenson and Hendry (2012). They distinguished five production planning and control stages in MTO and ETO companies: customer enquiry stage, design and engineering stage,
7
job entry stage, job release stage and shop floor dispatching stage. The first three stages can be considered to be part of the non-physical stage and will thus be relevant for this paper. An overview is given in figure 1, with the relevant part for this research in the box with the broken line. Only these three stages will be discussed in further detail.
Customer enquiry Design and
engineering Job entry Job release Shopfloor
Pre-production stage
FIGURE 1 – STAGES IN PRODUCTION OF ETO COMPANIES
The customer enquiry stage starts when a customer sends an invitation-to-tender or request for quotation.
This requires for example an estimation of lead times, assessment of the available design and production skills and an estimation of costs. These data are not always available during this stage of pre-production (Bertrand &
Muntslag, 1993). The second stage, the design and engineering stage, is where more detailed planning and design for already accepted orders takes place. This stage is of great relevance for ETO companies. During the design activity, the design of the product takes place while the engineering activities encompass the manufacturing of tools or machines that are needed for production. For this research, the design and engineering stage needs to be extended, as no test phase as been included in the model of Aslan et. al. (2012).
The expectation is that the type of companies this research focuses on will often use a validation stage like Phillips, Neailey & Broughton (1999) describe in their generic four-stage framework for product development.
As they will produce the product in high volumes and faulty products will lead to high costs. This stage encompasses the testing of a product before the actual production process begins, it will be added to the design and engineering stage as a new activity after the engineering activity in this research. After the design and engineering stage, the job entry stage begins. Production planning, material requirement planning and purchasing are all done in this stage.
It now seems like these stages are completely independent of each other and have clear borders, this will not be the case in practice. Decisions that have been made in the customer enquiry stage will affect all subsequent stages (Aslan et. al., 2012). When for example new orders are accepted in the customer enquiry phase, this might cause trouble in the design and engineering if they for example already face a high workload and do not have enough capacity. The other way around, the job entry stage will influence the customer enquiry phase as capacity constraints need to be taken into account to ensure that production due dates are feasible (Aslan et al., 2012).
Based on the above described stages and some examples of interdependencies a further defined model of the pre-production stage can be made. Figure 2 encompasses all activities within the stages of Aslan et. al. (2012) that are expected to be present in an ETO company. The description of specific activities will enable us to describe specific expected causal relations between uncertainty and these activities.
Request for quotation
Assessment technical capabilities and
costs
Design of the
product Engineering Validation Production
planning
Material requirements
planning
Customer enquiry Design and engineering Job entry
Material requirements
planning
FIGURE 2 – ACTIVITIES IN THE PRE-PRODUCTION STAGE
8
2.3 UNCERTAINTY
Uncertainty can be defined as the difference between the amount of information required to perform a task and the amount of information that is already available in the organization (Galbraith, 1973). In other words, more uncertainty will make it harder to fulfill a task in a proper way or maybe the task cannot be done at all.
Uncertainty will thus delay the processes in pre-production and cause a longer throughput time of the pre- production stage.
Uncertainty can be divided into two main categories: environmental uncertainty and system uncertainty.
Environmental uncertainty can be explained as the uncertainty that is due to uncertainty in demand. While system uncertainty is the uncertainty that is caused by the processes within the company (Ho, 1989). From the literature review, four relevant types of uncertainty could be distinguished within these two categories. These and the corresponding explanations can be found in table 1.
Type of uncertainty Explanation
Environmental uncertainty
Uncertainy of product specifications
(Bertrand &
Muntslag, 1993)
Uncertainty of product specifications (which can be both technical and non- technical specifications) is caused by the fact that in this stage parts of the product remain unknown. Decisions concerning lead-time, price and capacity have to be taken when still a lot about the exact design is unknown. When the engineering of the product progresses, more will be known about the exact specifications of the product.
Mix and volume uncertainty (Bertrand
& Muntslag, 1993)
Mix and volume uncertainty of future demand considers the uncertainty that comes with the forecast of production. Because products are often customer specific, it is difficult to make a detailed forecast. An extra cause of uncertainty is the moment of the order intake. A company often is unsure whether it will receive the order and at which point in time the order will be placed.
System uncertainty
Process uncertainty (Bertrand &
Muntslag, 1993)
Process uncertainty is caused by the fact that products are customer specific and new, it is very difficult to make an estimation on how much time and which resources are needed to produce and engineer the product, even if the design is already completely known.
Completion risks (Miller and Lessard, 2001)
Completion risks consider the engineering difficulties and novelty. This type of risk can be distinguished into two subtypes. Construction risk refers to the difficulties that may arise when actually building the project while operational risks refer to the possibility that future income flows will not materialize during the use of the product. This will not be applicable during this study.
TABLE 1 – TYPES OF UNCERTAINTY
Uncertainty of product specifications, mix and volume uncertainty and process uncertainty are the three types of uncertainty that Bertrand and Muntslag (1993) distinguished in their paper. The first two types can be traced back to the customer and can therefore be seen as environmental uncertainty. The latter one is caused by uncertainty in the company and is thus categorized as system uncertainty. Completion risk is one of the three risks that can be distinguished in large engineering projects (Miller & Lessard, 2001). The difference between uncertainty and risk is that risk can be described in statistical terms while uncertainty applies to situations of which the outcomes cannot be predicted (Miller & Lessard, 2001). However, these authors claim to also refer to uncertainty when they use the word ‘risk’. It can be expected that the subtype construction risks may be present in pre-production stages of ETO companies in particular. Companies might be unsure whether they are able to manufacture the product in accordance with the customer requirements. If not, this may lead to unhappy customers. The other two, less relevant, types of risk mentioned by Miller and Lessard (2001) are market related risks and institution risks. Their definition of market related risks looks a lot like the second type of uncertainty (mix and volume uncertainty) that was mentioned by Bertrand and Muntslag (1993). It refers to the uncertainty in demand and the difficulties with forecasting. Institution risks refer to the risk that comes with changing regulations and laws that might endanger the return on investment of a project. These risks are the most apparent in emerging economies, where the government could decide to renegotiate contracts or
9
influential pressure groups may arise. This seems to be a risk that is rather applicable for large construction projects and is less likely to be apparent in the environment this paper discusses. As the ETO companies this research focuses on do not have to deal with highly changing regulations or influential governments or pressure groups.
2.4 CONCEPTUAL MODEL
The pre-production stage has now been defined into the customer enquiry stage, design and engineering stage and a job entry stage. Within these stages, several activities which are expected to be present have been determined. In this part of the literature review, the conceptual model will be discussed.
Because all earlier mentioned types of uncertainty could be present and influential in all different activities in the pre-production stage, no specific expected relations will be described in this section. We will however, make a distinction between touch time and waiting time like Suri (2011) did. Touch time is the time employees actually work on the product, while waiting time is the time the product is still in the process but it is not being worked on. Every type of uncertainty can thus lead to both, extra touch time and extra waiting time, as shown in table 2. The actual research will focus on different types and causes for uncertainty in every activity of the pre-production stage, this will also be split out in the results section.
Type of uncertainty Influence on touch time Influence on waiting time Uncertainty of product specifications Process takes longer because
tasks need to be redone or are more difficult to complete.
Processes need to wait for extra information on the specifications and subsequent processes might need to wait because of delays in earlier processes.
Mix and volume uncertainty Processes might need to be redone depending on volumes that are ordered.
Processes need to wait for the confirmation of an order and subsequent processes might need to wait because of delays in earlier processes.
Process uncertainty Processes might take longer than planned.
Subsequent processes might need to wait because of delays in earlier processes.
Completion risks Processes need to be redone
when designs are not manufacturable.
Subsequent processes might need to wait because of delays in earlier processes.
TABLE 2 – INFLUENCE OF UNCERTAINTIES ON PRE-PRODUCTION
10
Request for quotation
Assessment technical capabilities and
costs
Design of the
product Engineering Validation Production
planning
Material requirements
planning
Customer enquiry Design and engineering Job entry
Purchasing Pre-production stage
Uncertainty of product specifications
Mix and volume
uncertainty Process uncertainty
Environmental uncertainty System uncertainty
Completion risks Uncertainty of
product specifications
Mix and volume
uncertainty Process uncertainty
Environmental uncertainty System uncertainty
Completion risks
Extra waiting time Extra touch time
FIGURE 3 – CONCEPTUAL MODEL
11
3. METHODOLOGY
The goal of this research is to further investigate uncertainty in the pre-production stage of ETO companies.
The development and origin of uncertainty in an organization will be investigated by doing an in-depth case study research. This chapter discusses this method and the ways of making sure that quality and reliability of the research is maintained. Firstly the method selection is discussed, then the selection of a case will be presented and then the data collection methods are discussed.
3.1 METHOD SELECTION
A single case study has been conducted to find an answer on the two questions that have been posed. A case study can be a very suitable type of research to explore phenomena that are poorly investigated so far.
Furthermore it is a good method to conduct an explanatory research and explain how a phenomenon works in a certain way in its environment (Yin, 2009). These are both issues which suit this paper very well. Three main advantages are mentioned in literature: (1) the phenomenon can be studied in its natural setting, (2) the method allows to answer questions with a relatively complete understanding of the nature and complexity of the phenomenon and (3) it is suitable for early and exploratory research in which the phenomenon is not understood yet (Meredith, 1998). These three advantages are all very applicable to this research. Firstly because very little research has been conducted on uncertainty in the pre-production stage of ETO companies.
Secondly because uncertainty needs to be studied in its natural setting as it is a complex phenomenon that can be influenced by many different factors in the direct environment.
3.2 CASE SELECTION
Only one case has been investigated during this study. The advantage of a single-case study is that a greater depth can be achieved. However, this also limits the generalizability of the conclusions and biases may occur more easily (Karlsson, 2009). A representative company in the metal forging industry has been found willing to participate in this research. The company is a leading company when it comes to forging in brass. It is a suitable company as it has a significant and clear pre-production stage with all relevant elements in it (customer enquiry stage, design and engineering and the job entry stage).
3.3 DATA COLLECTION
To get a good understanding of what is going on in the company of research concerning uncertainty it would be ideal to collect qualitative as well as quantitative data. The company however, could not provide much useful quantitative data for this research. They store some quantitative data for the actual production stage, but it is very limited for the pre-production part of the organization. Only some activity flowcharts of the departments in pre-production where used in this study as quantitative data.
Consequently, mainly qualitative data has been collected in order to answer the research questions that have been posed in the introduction. This is done by means of semi-structured interviews. In these interviews the focus was on the perceived uncertainty and the external and internal causes for the uncertainty and the perceived level of uncertainty in different departments. Employees from the entire production process have been interviewed and asked about their perception of uncertainty in pre-production and its causes during April and May 2014. Fourteen people from sales, design, planning, procurement, quality control, tooling, forging and machining have been asked for their opinion on this subject. This includes the managers of sales, design, planning, purchasing, forging and machining. Besides this one employee from sales, three employees from design, one employee from planning, an employee from forging and an employee from machining have been consulted.
12
An interview protocol with some guidelines when it comes to the questions (appendix A, in Dutch) has been used to give some structure to the interview and to make sure the interviews are as consistent as possible. This protocol is loosely based on parts of the case study protocol as suggested by Yin (2009).
These people have been made aware of the fact that all results from the interviews will be treated strictly confidential and that no information which enables the identification of the interviewees will be published.
Furthermore, it was pointed out that the recordings will only be used for transcribing and coding the interviews and no other people but the researcher will gain access to these tapes.
3.4 DATA ANALYSIS
Once the interviews in the company were conducted, these have been analyzed to retrieve information out of the data. The analysis of the transcripts of the interviews is done using ‘In Vivo Coding’. This type of coding refers to a word or small part of a sentence of the transcript. These words or phrases should seem to call for underlining or highlighting. So when something in the transcripts stands out, it should be applied as a code.
Such a code could for example be the phrase ‘uncertainty on specifications’. ‘In Vivo Coding’ is appropriate for all qualitative studies and well doable when the researcher has little experience with coding data (Saldaña, 2009). The ‘In Vivo Coding’ procedure provided us with a lot of different phrases which have been categorized using the four types of uncertainty as categories. We also kept track of the respondents who identified the uncertainties and the departments in which the specific uncertainties are present.
3.5 RESEARCH QUALITY
To assure the quality of a research, tests on construct validity, internal validity, external validity and reliability are often done (Yin, 2009). The first test is already a challenging one in case study research. Construct validity considers the measurement of the right concepts in the research. One tactic to attain the right level of construct validity is the use of multiple sources of data. The original aim was to retrieve both, qualitative and quantitative data to attain this validity. However, as quantitative data was not available, only qualitative data has been used during the research. This is partly compensated by the fact that fourteen different employees throughout the production process in the company have been interviewed. Thirteen of them have been used as a source for the research, as one interview has been rejected from the research because it did not yield any valuable results. The internal validity can be described by the ability to describe the right causal relationship, in this study for example a relationship between uncertainty and a specific cause. A first important step to maintaining internal validity is explaining what is meant with uncertainty in this research to the interviewees.
This has been done in the interview protocol (appendix A). Furthermore, the interviewer remained critical to the explanations about their perceived forms of uncertainty and their relations by often asking the question whether it was really uncertainty or if there could be another explanation. The third test, the one on external validity, considers the generalizability of the research. This is a common problem in case studies, especially in single-case studies. However, the company can be seen as representative for the specific type of companies this research has been conducted on. Generalizing the results for other companies within the ETO industry will raise more problems. The last test considers the reliability of the research. A reliable research can be repeated time and time again with the same results. The enable this, the procedures in this study need to be documented in such a way that anyone else could repeat the study. For this reason, and for the sake of consistency, an interview protocol has been used during the interviews (appendix A). Of course, this methodology chapter also provides helpful insights on how this research has been set up.
4. RESULTS
The findings of the research are presented in this chapter. The first section describes the company and their pre-production stage, the next section contains information about the information flows in the pre-production stage of the company and the results of the interviews are discussed in the last part of this chapter.
13
4.1 DESCRIPTION OF THE COMPANY
The company in which the study has taken place is a major company in the market of forging in brass in the Netherlands. After the pre-production stage, hundreds to millions of the same products are produced in the factory. Their products can be found in sanitary faucets, water meters, heating equipment, tubes, gas valves, beer dispense systems, trucks, cars or electrical devices and installations. Approximately one new order is accepted every week, apart from the repeat orders they receive. These repeat orders will not move through the pre-production stage again and are thus left out of this study. The company started with the implementation of Advanced Product Quality Planning (APQP) which is often required for customers in the automotive industry. APQP is used as a guide in the development process of a new product and makes use of standardized procedures to ensure the right quality and good communication between supplier and customer.
When checking with the characteristics of ETO companies, we can conclude that this company can be considered an ETO company. The production flow is completely driven by customer orders, customers have influence on production up to the design stage where designs that are delivered by customers are adjusted for manufacturing. This means that we are dealing with pure customization when it comes to new orders, as every product is developed completely new; partly by the customer and partly by the company.
New products first need to go through an extensive pre-production process which takes twelve weeks on average. The process is visualized in figure 4.
Customer enquiry
Design
Product Design tools Quality
Assurance
Purchasing tooling (- materials)
Sales
Purchasing
Design Calculation &
Routing
Product test
Planning Production
Planning
Purchasing Purchasing raw materials Production of
tools
Quality dep.
Tooling Design
FIGURE 4 – PRE-PRODUCTION STAGE IN THE RESEARCH COMPANY
The first phase is the sales phase, which is in accordance with the customer enquiry phase from the literature.
Then the the route through the factory and possible external activities is determined and the costs are calculated. Nesxt step is the design of the product, the drawings delivered by the customer are changed to meet demands of the production facilities of the company as the forging process creates some specific needs for the designs to be manufacturable. After the new designs are approved by the customer, the tools which are needed for production are designed. Dies need to be designed for the forging process, while drills and other tools are often needed in the machining part of production. The design phase is then finished. Quality assurance starts with their activities for the selection of suppliers and quality assurance planning, which is a planning in which the measures taken to assure the right quality are included. Purchasing will take care of the needed materials for the tooling department and buy the required externally produced tools. Note that the tooling department is called ‘engineering’ in the literature review of this study. When tooling finished the required tools, these will be tested in the test phase which is similar to the validation activity in the literature.
When these tests are successful, the design and engineering phase can be considered finished. Then a production planning is made and the purchasing of raw materials will start. The activities that are discussed here are the most common activities for a new product. It is needless to say that some activities will sometimes
14
be done at the same time or in a slightly different order, this will differ per product and will depend on the time left until the deadline. When there is much pressure on finishing the project, departments will try to start earlier than normally is the case or change the planning of other projects.
After these activities in the pre-production stage, production starts. The products are being forged and then go to the machining department where they are machined and sometimes even some assembling is being done.
4.2 REQUIRED INFORMATION PER DEPARTMENT
To find out what the input and output of information is for every department in the pre-production stage, two main information sources have been used. Activity flowcharts of several departments have been used as a start to get a grasp of the information flows in the company. After this, the list has been completed through conversations and discussions with the managers in the company. Table 2 gives an overview of the input and origin of the input in every department and the output and destination of the output of every department.
Department Input Output
What From What To
Sales Request for quotation Customer Request for calculation Design
Calculation Design Order confirmation Customer
Accepted quotation Customer Design order Design
Design
Request for calculation Sales Calculation Sales
Design order Sales Drawing package
(2D/3D/CAM)
Tooling
Drawing package Purchasing
Drawing package Production
Tool order Tooling
Routing Planning
BoM (piece level) Planning Required resources Planning Request for selection of
suppliers
Quality assurance Request for selection of
suppliers
Purchasing
Planning APQP Planning
Planning APQP Quality
assurance Measurement list FAI &
serial inspection
Quality assurance Measurement list FAI &
serial inspection
Production
Tooling Drawing package Design Tools Validation
Tool order Design Quality Assurance tools Validation
Validation
Tools Tooling Results validation (samples) Quality
assurance Quality Assurance tools Tooling Product information Design Quality Assurance planning Quality
assurance Planning resources Planning
15
Quality Assurance
Request for selection of suppliers
Design Rating selected suppliers Purchasing
Planning APQP Design Quality Assurance planning Validation
Results validation (samples) Validation Quality Assurance planning Customer Measurement list FAI &
serial inspection
Design PPAP (including samples) Customer
Planning Routing Design Planning validation Validaton
Required resources Design Production planning Production
Planning APQP Design BoM (order level) Purchasing
BoM (piece level) Design
Purchasing BoM (order level) Purchasing Purchasing external tools Supplier
Drawing package Design Order for raw materials Supplier
Request for selection of suppliers
Quality assurance
Order outsourced activities Supplier Rating selected suppliers Quality
assurance TABLE 3 – INPUT AND OUTPUT PER DEPARTMENT
4.3 INTERVIEW RESULTS
In order to acquire a sufficient amount of data, thirteen interviews across the company have been completed during the study. The causes for uncertainty during the pre-production stage have been categorized using ‘In Vivo Coding’ like described in section 3.4. Table 4 on the next two pages gives an overview of the results per specified per activity. The first column represents the department of the respondents, the second column shows the total number of respondents in each department, then the causes for uncertainty as explained by the interviewees are shown and the last columns show the corresponding type of uncertainty and their influence on the touch time and waiting time. After this, table 5 gives an idea of the frequency of every type of uncertainty per activity in the pre-production stage and the total number of times that the four types of uncertainties have been identified.
A global overview of the results is depicted in figure 5. It clearly shows that most uncertainty is present during the design and engineering phase and that most of this uncertainty is caused by uncertainty of product specifications, which already arises in the customer enquiry phase.
Department Number of respondents
Cause of uncertainty Place where
uncertainty is present
Type of
uncertainty
Extra touch time
Extra waiting time Sales 2 Too little time spent on customer requirements which causes uncertainty about
the specifications.
Request for quotation
Uncertainty of product
specifications
Yes Yes
You have to estimate up front whether you will be able to produce the design exactly like the specifications, especially in the automotive industry.
Assessment technical
capabilities and costs
Completion risks Yes Yes
Customers want to start with production to fast; before the engineering of a product is really finished. This often leads to changes of specifications from the customer, even during the design, engineering and validation activities.
Design,
engineering &
validation
Uncertainty of product
specifications
Yes Yes
Customers are not always completely clear about their specifications. What are the exact requirements, why are these requirements there and can these be changed a bit? These requirements and specifications also include things like packaging and ship sizes of the product.
Design, planning Uncertainty of product
specifications
Yes Yes
The manufacturing of tools causes uncertainty. More basic dies will take up to three weeks less to produce than specialized dies and we do not always know in advance which will be the case. You are never exactly sure how much time the manufacturing of tools will take.
Engineering Process uncertainty
Yes Yes
We sometimes accept orders that need a lot of redesigning before we are even able to produce it; because we want to score the order. This creates big risks for us and creates delays in the design phase of products, as designers have much extra work to do.
Design Uncertainty of
product specifications
Yes Yes
There always is some uncertainty concerning the timeliness of the validation of products. This needs to be planned in the production schedule, so the planners want to know this far in advance. However, we do not know when an order is confirmed so much time in advance.
Validation Mix and volume uncertainty
No Yes
Design and Engineering
4 Sometimes, our customer has not yet defined all specifications perfectly which leads to changes in specifications by the customer. This leads to extra work for the design department.
Design Uncertainty of
product specifications
Yes Yes
The planning in the tooling department is also very unsure. Sometimes, tools can be finished very quick. But sometimes it takes a few weeks longer because it is very busy in this department.
Engineering Mix and volume uncertainty
No Yes
For the validation it is the same as with the tooling department. When the machines are all planned for full capacity it might take longer and sometimes
Validation Mix and volume uncertainty
No Yes
17 we can shift some things in the planning.
We often get designs from customers which has not been worked out completely and of which we do not know the exact function. This leads to uncertainty for designers.
Design Uncertainty of
product specifications
Yes Yes
We also have uncertainty during the validation activities. It might sometimes take longer than what was planned because dies need to be changed a few times for example before the production is successful.
Validation Process
uncertainty
Yes Yes
Because the tooling department also has to deal with repeat orders, there workload differs and the time it takes to produce new tools differs, this is uncertainty we got to deal with as we only know at late point if the order is confirmed.
Engineering Mix and volume uncertainty
No Yes
We also get sometimes change requests from the customer after the design stage. This disturbs our processes and causes a lot of unexpected work.
Design,
engineering &
validation
Uncertainty of product
specifications
Yes Yes
We sometimes get designs with specifications that are not completely clear.
These things should be confirmed with the customer, but that does not always happen. This leads to uncertainty about designs.
Design Uncertainty of
product specifications
Yes Yes
It also seems that customer have not always worked through their own design very well. This leads to changes after and during the design activity on account of the customer.
Design, engineering, validation
Uncertainty of product
specifications
Yes Yes
Planning 2 There is some uncertainty in the planning of the validation stage. The moment of final order intake is very late in comparison to the periods I plan ahead with my production planning. So it is not always sure if there is time left in the production planning for validation.
Validation Mix and volume uncertainty
No Yes
We also often promise the customer a certain date they will receive their first series of products when we are still in the design phase. But when the validation phase goes wrong and we need to redesign and re-engineer, we get in trouble with this due date.
Design,
engineering &
validation
Process uncertainty
Yes Yes
There is often a lack of information when it comes to the non-technical aspects of products, for example about the shipping and the required amount of products per time period. This can be caused by the customer and ourselves as we do not ask enough for these things.
Production planning
Uncertainty of product
specifications
Yes Yes
Purchasing 1 Most uncertainty for purchasing comes from the late point in time that specifications are known or the fact that specifications are changed because they were wrong.
Purchasing Uncertainty of product
specifications
Yes Yes
Quality Assurance
1 What I see is that I miss information about what PPAP requirements a customer has. There is sometimes lacking information about due dates of the samples
Validation Uncertainty of product
No Yes
18
and likewise things. specifications
Production 3 What I sometimes see is that measurement programs for serial inspection still need to be made when production has already started. Which leads to work for design what already should have been finished and to uncertainty for us what needs to be measured and is acceptable.
Design Uncertainty of
product specifications
Yes Yes
We sometimes lack information about the precise tolerances of products, design is also sometimes unsure because you cannot speak through every single specification and its ultimate tolerances with the customer. This then needs to be communicated with the customer again which takes time of the designers.
Design Uncertainty of
product specifications
Yes Yes
We do often not know what is tolerable considering the tolerances and slightly crooked products. We do not know what is important and not important of a product and then ask the project leader which is the designer.
Design Uncertainty of
product specifications
Yes Yes
TABLE 4 – RESULTS PER DEPARTMENT
Frequency mentioned Activity Uncertainty of product
specifications
Mix and volume uncertainty
Process uncertainty
Completion risks
Request for quotation 1 0 0 0
Assessment technical capabilities and costs
0 0 0 1
Design of the product 11 0 1 0
Engineering 3 2 2 0
Validation 4 3 2 0
Production planning 2 0 0 0
Material requirements planning
0 0 0 0
Purchasing 1 0 0 0
Total 22 5 5 1
TABLE 5 – FREQUENCY OF UNCERTAINTY PER ACTIVITY
FIGURE 5 – UNCERTAINTY PER PHASE WITH FREQUENCIES IN BRACKETS Customer enquiry
phase
•Uncertainty of product specifications (1)
•Completions risks (1)
Design and engineering phase
•Uncertainty of product
specifications (18)
•mix and volume uncertainty (5)
•Process uncertainty (5)
Job entry phase
•Uncertainty of product specifications (3)
20
5. DISCUSSION
In this section, the two research questions which were posed in the introduction will be answered: (1) how can uncertainty in pre-production internally and externally be explained, and (2) during which activities in pre- production does uncertainty increase or decrease? The possible explanations for uncertainty in pre-production and the development of uncertainty during the activities as found in this case study will be discussed. This will be done in accordance with the structure of the pre-production stage: the customer enquiry, design and engineering and job entry phase will be discussed separately.
5.1 CUSTOMER ENQUIRY PHASE
The customer enquiry phase encompasses the request for quotation and the assessment of technical capabilities and costs. Regarding the request for quotation activity, this has only be mentioned by one employee as an activity where uncertainty is present.
An employee from the sales department stated that they do not always spend enough time on the customer requirements because they want to score the order. This leads to incomplete information about the product that needs to be changed during later activities in pre-production, an example of uncertainty of product specifications. Because of this uncertainty, processes take longer and need to be redone leading to extra touch time and longer waiting time for the subsequent processes. The sales employee also indicated that they are not always sure whether the production of a design is technically possible for the company during the assessment of technical capabilities and costs. This could lead to extra communication and changed designs which, in its turn, leads to extra touch time and extra waiting time.
Based on these conclusions from the interviews, one could say there is not much uncertainty during the customer enquiry phase. This is probably not true. A lot of respondents stated that there is much uncertainty of product specifications, especially during the design and engineering phase (18 times when the three activities during this phase are summed up). This uncertainty of product specifications can, for a large part, be assigned to the customer enquiry phase as this is the phase where these specifications should become clear. It however does probably not have many consequences as the sales people do not have to work with the information they get. But when the designers start working, all kinds of difficulties rise. This means that this kind of uncertainty does not come to surface before the design activity in the company, however a large part of it is already caused during the customer enquiry phase. Furthermore, the moment of order intake (mix and volume uncertainty) has been pointed out by several employees across the departments as a troublemaker in the engineering and validation stage. Obviously, this is also uncertainty that is caused during the customer enquiry phase (moment of order intake) but has consequences for engineering and validation.
Concluding, we can say that most uncertainty (uncertainty of product specifications and mix and volume uncertainty) is caused during the customer enquiry phase. This has also been recognized by two sales people that were interviewed during this study. They stated that they sometimes have the tendency to accept orders too fast, without having all requirements of the customers clear and spend too little time on the communication with customers. But one sales employee also stated that customers sometimes push too hard to get their product in production very quickly, before the product is matured completely. It seems that uncertainty of product specifications is not a purely environmental factor as suggested in the theoretical part.
Even employees at the sales department state that they are sometimes too eager to confirm a new order, without ruling out as much uncertainty as possible. This makes it partly an internal factor, which means that it can also be seen as system uncertainty.
