Master Thesis: An outsourcing guide for Philips grooming
“A case study on the outsource decision of the electronic design of grooming products”
By Bert Emmelkamp
Appingedam, April 2011
University of Groningen
Master Thesis: An outsourcing guide for Philips grooming
“A case study on the outsource decision of the electronic design of grooming products”
Author: ing. B.B. Emmelkamp Student number: 1733524
Address: Harm Kuperstraat 21 9901 GT Appingedam Telephone: 06-18959677
E-mail: bert.emmelkamp@student.rug.nl Period: Sept. 2010 - Apr. 2011
1st supervisor ing. A. Tuinstra
Address: Philips Consumer Lifestyle Oliemolenstraat 5 9203 ZN Drachten Telephone: 0512-592397
E-mail: auke.tuinstra@philips.com
1st supervisor: Dr. Ir. M.W. Hillen 2st supervisor: Dr. Ir. N.R. Faber
Address: Faculty Economics and Business University of Groningen Nettelbosje 9747 AE Groningen Telephone: 050-3638966
Page. 3
Management Summary
This research concerns the outsourcing decision of the electronic design (PCB design) of grooming products by Philips Consumer Lifestyle in Drachten. The development of the PCB designs currently takes place in-house. However, the initial idea was to outsource this development to subcontractors in order to reduce costs and time. Although Philips has already tried to outsource the PCB design of its grooming products, it appeared that in many cases, problems arose and it turned out to not to be profitable at all. An extensive internal and external analysis to define the problem statement was conducted to find the actual underlying problems of this situation. Based on this problem statement an analysis and diagnosis was carried out.
During this analysis and diagnosis phase various analyses were carried out. First the PCB design outsourcing process was researched. It appeared that during this process several problems occurred in supplier involvement, supplier responsibilities and disruptions in the communication process. After the PCB design outsourcing process was clear, the supplier involvement was scrutinized. During this study it became clear that in some cases it is better to involve suppliers in an earlier stage of the project than involving them halfway. Moreover, a way to determine when suppliers must be involved in the NPD process was found. After that, the supplier responsibilities were investigated. It appeared that Philips always takes full responsibility, which involves higher expenses as well as an extra investment of time. During this analysis a way to transfer more responsibility to the supplier was investigated. Therefore, Philips needs to change from a lower intensive supplier collaboration (white box) towards a higher intensive supplier collaboration (gray box and black box). During these forms more responsibilities are shifted to the supplier, to ultimately make them fully responsible for the design. Finally the communication process was investigated. It appeared that some ambiguities (wrong interpretations) existed between Philips and their suppliers. Therefore, ways to overcome these ambiguities were scrutinized, which were mainly based on more face-to-face contact.
case of LE PCB designs halfway (stage 3 to 6) the NPD process, and to involve them during the whole NPD process (stage 1 to 6) in case of ME and HE PCB designs.
The final step presents recommendations to transfer more responsibility for the PCB design to the supplier by moving from a white box design to a grey box design. Once a grey box design proves successful, a black box design is highly recommended to be the last step. Even though suppliers have not matured to this stage yet, supplier development programs can help them become more capable of using a grey or black box design. A maturity grid points out that supplier 5 seems the best supplier to select for starting with these recommendations. Finally, the results of this research can also be generalized so that they are also applicable to the product categories Shaving and Vita-light.
Page. 5
Preface
This Master thesis describes the final research project which I conducted as part of the Master’s program in Technology Management at the faculty of Economics and Business of the University of Groningen. I will start with a short introduction of my prior school career and my other activities.
For me it all began at the age of 6 at the primary school and after that I went to the Dollard College in Woldendorp where I obtained my MAVO diploma at D-level. After graduation, I went to the Noorderpoort College in Appingedam/Groningen to study MBO electronics. After I had completed this program, I knew that I really wanted to learn more about electronics, so I went to the Hanze University of Applied Sciences in Groningen to study HBO Electronic Product Design and Engineering (EPDE). After graduation, I was ready to go and find a job, but I wanted to broaden my horizon and looked for a study that combined my technical knowledge with business knowledge, which for me was a whole new field to learn. I therefore chose to study Technology Management at the University of Groningen. I started with a pre-master Technology Management program. After receiving my Bachelor of Science degree, I began with the final master Technology Management program. Besides my educational activities, I also run an electronic engineering company called J&B Systems together with a friend. This company works exclusively for the Dutch telephone company KPN. In my company, I can combine my electronic skills with the business skills I learned at university. At the age of 26 and after spending over 20 years at different school desks, I can finally put what I have learned into practice.
This master thesis is the last assignment before obtaining my Master of Science degree. I carried out the final research project which is the basis for this Master thesis at the Electronics & Software Engineering department of Philips Drachten. During an internship that lasted 8 months, I investigated the outsourcing decisions of the electronic design of grooming products. With this project I combined my electronic background with my business skills. I want to thank the people who helped me finish this Master thesis successfully. First of all, I would like to thank my company supervisor Auke Tuinstra for his guidance, comments and support during my research. I also would like to thank my company director Eric Bosscher for giving me the opportunity to do my internship within Philips. I would also thank all my other colleagues for helping me with this research project. I would like to thank my first supervisor Michiel Hillen for all his comments, his guidance and support. I would also thank my second supervisor Niels Faber for his support and for reviewing my Master thesis. Finally, I would like to thank my parents and my lovely girlfriend Ineke for their care and support. During this graduating project they helped me with motivation and inspiration.
Index
Management Summary ...3 Preface...5 Index...6 List of abbreviations ...8 1. Introduction ...9 1.1 Motivation ... 91.2 The company of research ... 9
1.3 The product ... 10
1.4 Supplier relationship... 11
1.5 Management question ... 13
1.6 Typifying the research ... 13
1.7 Reader guide... 14
2. Problem identification ...16
2.1 The organization as a system... 16
2.2 Internal analysis ... 16
2.3 External analysis ... 18
2.4 The problem explication ... 18
2.5 Conceptual causal model... 22
2.6 Problem statement & Research questions ... 23
3. Methodology...25
3.1 Data collection ... 25
4. Analysis and diagnosis ...26
4.1 Background information ... 26
4.2 The PCB design outsourcing process ... 27
4.3 Supplier involvement... 29
4.4 Supplier responsibilities... 31
Page. 7
5. Plan of action and conclusions ...38
5.1 Summary of the recommendations ... 38
5.2 Explanation of the recommendations ... 40
5.3 Maturity grid... 44
5.4 Generalizing the solutions ... 45
6. Research limitations and further research...46
6. 1 Research limitations ... 46
6. 2 Further research ... 46
Appendix I: Background information ...52
Appendix II: Internal interview (Philips) ...54
Appendix III: Qualitative data ...56
Appendix IV: External interview (OEM/EMS supplier) ...59
Appendix V: Benefits of ESI...62
Appendix VI: NPD process ...64
Appendix VII: PCB Design outsourcing process ...67
Appendix VIII: Questionnaire...70
Appendix IX: Supplier involvement, responsibilities and communication ...74
Appendix X: Supplier development grooming ...77
List of abbreviations
ARC Application Research BOM Bill Of Materials
BPS Business Problem Solving CAD Computer Aided Design CCM Conceptual Causal Model
CE Consumer Electronics
CL Consumer Lifestyle
DAP Domestic Appliances and Personal Care DDC Diagnosis Design Change
DFM Design For Manufacturing EMC Electronic Magnetic Compatibility EMS Electronic Manufacturing Service
EPDE Electronic Product Design and Engineering ESE Electronics & Software Engineering ESI Early Supplier Involvement
FOS First Official Samples
FOT First Of Tools
HE High End
IDA Innovation Domestic Appliances IPC Innovation Personal Care IPD Integrated Product Development IT Information Technology
LE Low End
ME Medium End
NDA Non Disclosure Agreement NPD New Product Development OEM Original Equipment Manufacturer PCB Printed Circuit Board
PCBA Printed Circuit Board Assembled R&D Research and Development
Page. 9
1. Introduction
The first section describes the motivation, which is the reason behind this research. The second section gives information about the company at which the research took place. The third section discusses the product this research is about and the main parts of it. This section also discusses a small part of the problem. The fourth section discusses the supplier relationship between Philips and their suppliers. The management question and the typifying of this research are elaborated on in the fifth and sixth section respectively. The seventh section contains a reader guide that explains the structure of this research.
1.1 Motivation
When a new product is made, Philips can choose to develop certain parts in-house or to outsource the development of those parts to a subcontractor. The development of the electronic design of grooming products is currently being done in-house at Philips Consumer Lifestyle in Drachten, but the future ambition is to outsource this to subcontractors in order to save costs and time, especially since the manufacturing of the complete product already takes place in foreign countries and the development of other parts of the product (cutting element, housing, etc.) is increasingly being outsourced too. It therefore seems a logical step to outsource the electronic design of grooming products as well. In addition to this, Ernst and Kamrad (2000) argue that approximately 80% of the manufacturing cost of a product is determined by the design of the product. The opportunities for further savings lie in the integration of product design (Mikkola and Skjoett-Larsen, 2003). A way for Philips to achieve this is to involve their subcontractor early in the new product development process.
Previous projects had shown that Philips was worse off when they decided to outsource electronic design to a subcontractor. There were many problems and Philips didn’t know how they had to make the right outsourcing decision. This research aims to get more insight into the issues that arise during this outsourcing decision and about how to deal with these decisions. Although most literature is about outsourcing in general (mostly production), outsourcing of New Product Development (NPD) is relatively new and academic research is limited in this field (McCutcheon et al., 1997). Mikkola and Skjoett-Larsen (2003) also state that very little academic research has been done about the collaboration between manufacturers and their suppliers with respect to NPD. This research can therefore deliver practical insights, but at a theoretical level there appears to be a lot to investigate yet. The next section will continue by giving some background information on the company where this research took place.
1.2 The company of research
delivers lifestyle solutions for the personal wellbeing. The biggest site of the CL sector is located in Drachten. On this site the well-known Philips shaver is developed and produced. This site also develops a range of other consumer lifestyle products, such as Wake-up lights, Senseo coffee machines, vacuum cleaners, groomers, etc. At Philips Drachten, the innovation site ‘Innovation Personal Care’ (IPC) is mainly responsible for the development and creation of the following three product portfolios:
• Shaving (e.g. Philips arcitec, SpeedXL) • Vita-Light (e.g. Wake-up Light) • Grooming (e.g. Beard trimmer)
This research was conducted at Philips IPC at the Electronics & Software Engineering (ESE) department within the product category Grooming. This department deals with the electronic and software development for grooming products. The products that this product category develops are products, such as beard trimmers, hair clippers, nose and ear trimmers and body groomers. Figure 1 shows some examples of grooming products within this range. In appendix I more background information about Philips as an organization, the CL division and the department of research can be found.
Figure 1: Product range of the category Grooming.
1.3 The product
The product category Grooming develops groomers that roughly consist of three parts. These parts are: a housing, a cutter and an electronic design (power supply). This electronic design is a small printed circuit board (PCB). This PCB contains functions to drive the motor, to charge and measure the battery, to adjust the device using buttons and to indicate information to the user using LED’s. Figure 2 shows an example of the top and bottom of a PCB that is being used in a grooming product. Figure 3 shows an example of a Computer Aided Design (CAD) drawing of a Printed Circuit Board Assembled (PCBA) which already has components on the board.
Page. 11 An electronic design in this research means a schematic and/or PCB layout. For the sake of simplicity, the electronic design will be referred to as a ‘PCB design’ in this research. Each new PCB design is classified by the acronym ‘HE, ME and LE’. These acronyms stand for High End, Mid End and Low End, and are used to serve different market segments. This classification specifies the amount of functions a design has, so an HE design has more functions than an ME design or an LE design. These three classifications most of the times consist of the same PCB, but LE designs have fewer components than ME or HE designs because they have fewer functions. ME and HE designs look quite similar with respect to functionality and design most of the times.
During the development of the PCB design for a new groomer, the ESE department can choose to do the PCB design in-house or to outsource the PCB design to a subcontractor. As can be read in appendix I, the ESE department started in 2008 with only one person who was responsible for the PCB designs. Today more than four persons ware working in the department, handling 90% of all PCB designs. The other 10% of the PCB designs are outsourced, a process that currently has a very explorative character. The reason that only 10% of the PCB designs are outsourced is that the outsourcing decision concerning PCB design is unclear. It appears that there are many problems around this issue. If it is decided to outsource a PCB design, the ESE department can choose between different options, can do business with different subcontractors, and have to deal with many flows between its subcontractors. This is discussed in the next section.
1.4 Supplier relationship
When it outsources grooming products, the ESE department makes use of subcontractors, also called first-tier suppliers. First-tier suppliers directly supply to the assembler/buyer or have a significant technical influence on the assembly when supplying indirectly. The Grooming category does business with two kinds of first-tier suppliers. These are an Original Equipment Manufacturer (OEM) and an Electronic Manufacturing Service (EMS) supplier. An OEM manufactures products or components that are purchased by a company and retailed under the purchasing company's brand name, which is in this case Philips. EMS suppliers are companies that design, assemble, produce, and test electronic components and do the PCB assemblies for OEMs. Table 1 shows the different OEM’s and EMS suppliers that the ESE department is using.
OEM (Censored) EMS Supplier (Censored)
Supplier 1 (Hungary) Supplier 5 (China) Supplier 2 (China)
Supplier 3 (China) Supplier 4 (China)
Table 1: OEM’s and EMS suppliers that the ESE department is using.
Currently there are three options with respect to outsourcing a PCB design:
The EMS supplier manufactures the complete PCB’s and also places the components. After this, the PCB’s are called PCBA’s. The EMS supplier then sends the PCBA’s to the OEM and the OEM finalizes the grooming product by combining the PCBA’s with the other components (housing, cutter, etc.). 2. The second option is to outsource the PCB design to an OEM or in some cases to an EMS supplier
(currently this is quite rare, so in the rest of this thesis only the OEM is mentioned when it comes to outsourcing a design). In this case, the particular subcontractor (OEM or EMS supplier) makes the PCB design. After the PCB design is made, the ESE department reviews their PCB design. Based on this review, they make changes until the ESE department accepts the PCB design. Then the PCB design is sent to an EMS supplier that will produce the PCB’s and transforms them into PCBA’s. The EMS supplier then sends the PCBA’s to the OEM that finalizes the grooming product by combining the PCBA’s with the other components.
3. The third option is just like the second option, with the exception that the PCB design is not accepted by the ESE department after several reviews. In this case the ESE department takes over the design.
From the above mentioned three options one can see that there are many flows between the ESE department, the OEM and the EMS supplier. These flows consist of an information flow and a material flow. The information flow stands for the communications between the different parties. But this flow also contains the PCB design, since the PCB design does not consist of any physical material at all. The material flow stands for the physical products, such as prototypes, samples and final products. Figure 4 shows the relations between the three parties.
Figure 4: The relations between ESE department, OEM and EMS supplier.
Page. 13
1.5 Management question
In this section the original complaint is stated in the form of a management question which is derived from the first intake meeting and several in-depth discussions with key people working in different fields. The intake meeting was held with the director innovation & development of the ESE department (Eric Bosscher), the head of Electronics and Software Engineering for grooming products (Auke Tuinstra) and with the overall architect (Oedilius Bisschop). The first thing that became clear from the intake meeting was that in some cases the ESE department is worse off when they decide to outsource the PCB design to an OEM. Sometimes the ESE department takes the responsibility of projects back, even though the PCB designs were already outsourced to an OEM, which leads to extra costs. According to the head of Electronics and Software Engineering for grooming products (Auke Tuinstra), the responsibility between the different parties is currently not clear. Another problem he mentioned was that it is not clear when to involve the supplier in the product development process. According to the head of the ESE department (Eric Bosscher) some subcontractors are not mature enough in product development and that they have a passive attitude in their way of working. According to him, collaboration with these subcontractors disrupts the development process and leads to rework. Rework leads to more costs and delays (longer throughput times) and lower business results, according to the head of the innovation team grooming (Peter Werkman). In addition to this, the PCB designs of previous projects were not prescribed. . The consequences of this decision were extra costs, bad quality and more risk, according to the head of the purchasing department (Laurent Hery). He stated that this was the reason the PCB designs of all outsourced projects are now prescribed, so that the ESE department is in charge of the electronic design.
This is the reason that the ESE department wants to know the correct method for outsourcing in the future. From the above it is obvious that there are two aspects that require investigation. The first aspect to be investigated is when to involve suppliers in the NPD process. Currently this is based on past experiences from previous projects. The second aspect to be examined is how the responsibilities of the different parties (the ESE department, OEM and EMS supplier) are divided in the case of outsourcing the PCB design. This is because the responsibility is currently not clearly defined and the ESE department wants to know their role in this process. The goal stated during this intake meeting was to get shorter throughput times and lower costs for the PCB designs. The next section discusses the kind of methodology used for this research.
1.6 Typifying the research
Figure 5: The regulative cycle (van Strien, 1997).
This cycle consist of, 1) the problem mess, 2) problem definition, 3) analysis and diagnosis, 4) plan of action, 5) intervention, 6) and an evaluation. These six steps are analogous to the diagnosis and design phase of the DCC model of De Leeuw (2002), as one can see in the figure. However, due time limitations and the scope of this research, only steps 1 to 4 are used. The model of van Aken et al. (2007) is mainly used for this research because it has more specified steps within each phase and because the problem has a BPS character (since the aim is to improve the performance of the business). The model of De Leeuw is also partly used to divide the six steps of the regulative cycle into the diagnosis, design and change phase (see section 1.7). Van Aken et al. (2007) and De Leeuw (2002) emphasize the importance of employing a pluriform diagnosis and a proper problem diagnosis. This research will therefore also look outside the organization of Philips to get a better picture of the actual situation. Section 3.1 in chapter 3 discusses how the data collection in this research took place. The following section describes the structure of this research.
1.7 Reader guide
Figure 6 shows a reader guide which serves as an overview of the structure of this research.
2. Problem identification
In the first section the organization as a system is defined to understand the scope of the problem. In the second section an internal analysis will be done to look further than the given management question. This analysis will consist of qualitative and quantitative data. To look further than the organization itself, also an external analysis will be done in the third section. The information that is collected from all these information sources results in a so-called ‘problem explication, which is covered in the fourth section. From this problem explication, a conceptual model is derived in the fifth section. The sixth section gives the problem statement and the research questions which are leading for this research.
2.1 The organization as a system
According to the system theory thinking, an organization can be seen as a system with its properties and an output (Boulding, 1956; De Leeuw, 2002). To understand the scope of the problem, it is important to define the system and its output. As this research took place at Philips Personal Care in Drachten it is considered as the system in this problem diagnosis. The output is the launch of successful and profitable products on the consumer market in accordance with Philips CL’s strategy. In this system, there is the ESE department where this research takes place. Figure 7 shows Philips as a system and its output.
Figure 7: The Philips organization as a system and its output.
According to van Aken et al. (2007), a problem can be seen as the result of a certain perception of state of affairs in the real world with which one or more important stakeholders are dissatisfied. Figure 8 shows a systematic view of the functional problem and the process that has caused it.
Figure 8: Systematic view of the problem as the output of a process.
2.2 Internal analysis
Page. 17 the business problem was discussed with important stakeholders in the problem. The first subsection will give qualitative data which was obtained by holding interviews with different employees that were dealing with the problem. According to van Aken et al. (2007), it is also important to analyze quantitative data. Therefore the second subsection gives an overview of the secondary data of earlier projects.
2.2.1 Qualitative data
To get a better understanding of the actual problems that exist, interviews were held with employees of the ESE department that are working in different fields (like engineering, purchasing, etc). By holding these interviews, a more valid picture from different angles is presented, which gives more insight. The interview and all the answers can be found in appendix II. The following conclusions can be deducted from these interviews:
• The PCB design capabilities of most OEM’s are limited because most OEM’s produce standard PCB designs.
• The responsibility between the ESE department, the OEM and the EMS supplier is currently not clear. In some cases the ESE department takes full responsibility for designing the PCB’s. This lack of clear responsibility leads to a disruption of the communication process.
• There is too much unnecessary communication because of different expectations and perceptions. Apart from giving technical support, suppliers’ planning issues are being dealt with.
• In some cases in which the PCB design is outsourced, a lot of revisions (rework) were made to make the PCB design acceptable, which lead to extra throughput times because of delays in the designing process.
• The changing process has a rough character (not systematic).
• Suppliers have difficulties understanding DFM issues during design reviews. This is mainly caused by cultural differences.
2.2.2 Quantitative data
Quantitative data was collected from multiple spreadsheet files which showed how many hours the ESE department had spent on previous projects. From these data two cases were derived for further investigation. All cases followed a product development process of roughly five stages. These stages are: ‘Assignment preparation, Project confirmation, Product implementation, Process verification & Initial production and Mass volume Ramp-up’. In these two cases the designs of some projects were outsourced while other projects were not. Appendix III contains the complete results of these two cases.
Conclusion case 1: Project X compared to Project Y (Censored)
the third stage in project Y, against only 482 hour in project X. According to the project leader of project Y, Marten van Wije, the initial decision to outsource the PCB design was taken with a view to reducing the hours of the ESE department. The ESE department took the PCB design back because of the bad quality the OEM delivered. Concluding one can say that the project that was being outsourced (and that failed) cost a factor of 2.4 more in total R&D hours than the project that was done in-house. Also with respect to project time there is a significant difference of 3 months extra time for the project that was outsourced.
Conclusion case 2: Project X compared to Project Y (Censored)
The whole PCB design (schematic & PCB layout) of project X was done by the ESE department, whereas the whole PCB design of project Y was outsourced to an OEM. Because the OEM had problems with the concept in stage 2 and because they had problems making a proper prototype in stage 3 due to their limited technical capabilities, they needed extra time. With respect to the total project time, there is a difference of 5 months extra time for the project that was outsourced.
2.3 External analysis
Besides an internal analysis also an external analysis was conducted to get the insights from the OEM and EMS supplier during the outsourcing of a PCB design. At these companies, semi-structured interviews were held with employees working in different fields, such as engineering, project management, etc. The interview and all the answers can be found in appendix IV. The following conclusions can be deducted from these interviews:
• Normally, the OEM should take the responsibility when deadlines are not reached. However, in some cases, for instance when technical issues cannot be solved, the ESE department takes over their responsibility.
• Engineers of the ESE department involve suppliers in their design reviews (DFM) too little.
• There are to some extent different expectations and perceptions with respect to the communication process between the ESE department and the OEM and EMS supplier.
2.4 The problem explication
Page. 19 this diagram. Figure 9 contains the results of the fishbone diagram. The four main categories of causes used for this research are: Method, Management, People and Measurement. These will now be further explained.
Method
The process of making changes to the PCB design when it is outsourced to an OEM has a rough character (not systematic) most of the times. A consequence of this is that the PCB design is changed constantly. In some cases a lot of revision (rework) is done to make the design acceptable. Another thing that occurs is late project changes due to the fact that suppliers are not mature enough and because of the fact that currently no roles and responsibilities are defined between the ESE department and its suppliers.
Management
It is currently not clear to which level the supplier must be involved in the NPD process, as there are no clear outsourcing criteria yet. Another point that is unclear is the division of responsibility between the ESE department and its suppliers because of the absence of clearly defined roles. It appears that the ESE department always takes full responsibility when the PCB design is outsourced. Besides only technical support, the ESE department also takes responsibility for planning and progress issues.
People
The communication between the ESE department and their suppliers is of a rough character (not systematic) and, because of that, too much (unnecessary) communication takes place. The ESE department and its suppliers also have different expectations about each other due to different perceptions and interpretations. No roles and responsibilities have been defined between the ESE department and their suppliers. These subcauses lead to the disruption of the communication process between the ESE department and their suppliers. Another subcause is the passive attitude of suppliers. Due to this attitude, they do not actively intervene in the NPD process when things go wrong.
Measurement
Management
Measurement People
Method
Unclear outsourcing decision concerning PCB design
Limited design capabilities of suppliers
Suppliers are not mature enough in developing new technology PCB designs
Suppliers disrupt the development process Disruption of the
communication process
Communication is rough (not systematic) which leads to too much communicatiom Rework by changing the design constantly
Process of changing the design is rough (not systematic) when it is outsourced
No roles and responsibilities defined between Philips and suppliers Outsource criteria unclear Late project changes Suppliers have a passive attitude Suppliers are not mature enough in developing new technology PCB designs Suppliers not sufficient involved in design reviews (DFM) DFM issues for suppliers unclear Philips and suppliers have
different expectations about each other and there are no roles and responsibilities defined
Supplier responsibility in the NPD process is unclear Supplier involvement in the NPD process is unclear No roles and responsibilities defined between Philips and suppliers
Figure 9: Fishbone diagram derived from the problem explication.
Normally all main causes in the fishbone diagram need to be empirically tested in order to check if they really exist. However, due to time limitations, a shortcut will be made in order to determine what the root causes are. In order to make this shortcut, respondents that are working in different fields were asked to rank each of the main causes from the fishbone diagram. All respondents are directly involved in the problems mentioned in the fishbone diagram. Table 2 shows the list of respondents.
Respondents (N=6)
1. Auke Tuinstra (Head Electronics & Software Engineering Grooming) 2. Deqiang Guo (Initial buyer)
3. Eric Bosscher (Director Innovation & Development ESE) 4. Jeroen Nijdam (Electronic engineer)
5. Luigi Balestrini (Electronic engineer) 6. Oedilius Bisschop (Overall architect)
Table 2: Respondents.
Page. 21
Main causes Mean Std. Dev
Method
1. Late project changes 4.8 1.7 2. Rework by changing the design constantly 5.3 2.1
Management
3. Supplier involvement in the NPD process is unclear 7.2 2.3 ← 4. Supplier responsibility in the NPD process is unclear 7.0 2.8 ←
People
5. Disrupted communication process 8.0 1.4 ← 6. Suppliers have a passive attitude 4.8 2.1
Measurement
7. Limited design capabilities of suppliers 6.7 1.5 8. Suppliers disrupt the development process 5.5 2.1 9. DFM issues for suppliers unclear 4.0 1.8
Table 3: Results of determining the root causes of the fishbone diagram.
The table shows that the three most important root causes are numbers 3, 4 and 5 because they have the highest mean value. These root causes are also indicated by the three arrows next to the table. The first root two root causes are the supplier involvement and the supplier responsibility during the NPD process. The third root cause is the communication process that seems to be disrupted. All three root causes were also found several times in literature. Liker et al. (1996) confirms this by saying that issues that arise with regard to supplier integration include extent and timing of supplier involvement, design responsibility, and inter-company communication. These three root causes are therefore chosen because they are the result of the ranking process and because they are mentioned several times in literature. It seems that these three root causes have therefore some importance with respect to the problem. Figure 10 shows these root causes in a Conceptual Causal Model (CCM) which shows their relationships to the functional problem. All three root causes and their relationship within the CCM will now be further elaborated on.
1. Supplier involvement in the NPD process is unclear
2. Supplier responsibility in the NPD process is unclear
It is not clear what the responsibilities of the suppliers are because of the absence of clearly defined roles. Normally the supplier should take the responsibility when a PCB design is outsourced to them. In some cases, however, the ESE department always retains full responsibility. Supplier responsibilities are currently based on past experience with previous projects. This method does not lead to a decision with a strong foundation. In addition to this, Petersen (2005) argued that the supplier’s level of responsibility during the NPD process is an important issue. Therefore it seems important to investigate what the responsibilities of the supplier are during the NPD process. This node in the CCM has a positive relationship with the functional problem because it states that the unclearer the supplier involvement is, the unclearer the outsourcing decision concerning PCB design becomes.
3. Disruption of the communication process
It appears that there is a lot of communication going on between the ESE department and its suppliers. This has to do with the different perceptions and interpretations between both parties, but also because no roles and responsibilities have been defined. Sometimes it appears that information is not understood by the other party or is misinterpreted. Because of this, the communication process is disrupted. When suppliers are involved, managing intra-organizational communication becomes an important NPD task, according to Hartley et al. (1997). Communication should also improve the quality of the NPD process, because it can reduce uncertainty and equivocality (Hartley et al., 1997). It seems therefore important to investigate the current communication process and how this can be improved. This node in the CCM has a positive relationship with the functional problem that states that the more the communication process is disrupted, the more the outsourcing decision concerning PCB design becomes unclear.
2.5 Conceptual causal model
Page. 23
Figure 10: Conceptual causal model of the problem.
2.6 Problem statement & Research questions
In this section the goal of this research will be stated in the form of an objective. After that a main question and a few subquestions will be formulated. During the intake meeting (see section 1.5) an objective was stated. The objective of this research is stated as follows:
Objective:
• “Shorten the throughput time and lower the costs of the electronic design for grooming products.”
In order to achieve this objective, it must be clear how the ESE department has to make the right outsourcing decision in order to have lower throughput times and lower costs regarding PCB designs. The following main question will therefore be central to this research and it is stated as follows:
Main question:
• “How should the ESE department improve their PCB design outsourcing process to make a clear outsourcing decision, which leads to a reduction in costs and throughput time?”
To answer the main question, several examinable subquestions are formulated. In order to improve the PCB design outsourcing process, the current PCB design outsourcing process needs to be investigated first.
Subquestions:
Therefore the first subquestion is:
1a. What is the current PCB design outsourcing process?
After the PCB design outsourcing process is clear, the current level of involvement of suppliers during the PCB design outsourcing process must be clear. The second sub question is:
2a. What is the current level of involvement of suppliers in the PCB design outsourcing process? 2b. How can the level of involvement of suppliers in the PCB design outsourcing process be improved?
Besides the level of involvement, the responsibility of the suppliers during the PCB design outsourcing process must be clear. Therefore the third question is:
3a. What are the current responsibilities of suppliers in the PCB design outsourcing process? 3b. How can the responsibilities of suppliers in the PCB design outsourcing process be improved?
There is a lot communication going on between Philips and its suppliers because of different expectations and interpretations of both parties. Because of this, the communication process is disrupted and needs to be improved. Therefore the fourth subquestion is:
4a. What is the current communication process during the PCB design outsourcing process between the ESE department and its suppliers?
Page. 25
3. Methodology
In section 1.6 (typifying the research) the type of methodology used for this research has already been elaborated on. The reason to state the methodology at the beginning of this research was to give the reader an idea of the methodological steps in this research. The model of van Aken et al. (2007) is mainly used as a guide for this research. During this research, data needed to be collected as well. The following section discusses how this data collection method took place.
3.1 Data collection
This section discusses the method for collecting data. As was made clear in section 1.3, there is still not much information available on the outsourcing of PCB designs, because outsourcing currently has a very explorative character. Therefore, most data had to be collected by interviews and in-depth discussions with employees of the ESE department. Eisenhardt (1989) argues that multiple data collection methods and sources provide stronger substantiation of constructs and hypotheses, strengthening convergence of results. The data for this research are therefore based on primary and secondary data. Primary data that were observed or collected in the organization were derived by semi-structured interviews with employees in the organization and the external organization. The reason to interview in a semi-structured way was that employees were able to freely express their views and experiences on the outsourcing decision. According to Eisenhardt (1989), obtaining data from multiple informants increases the quality of the collected data. Therefore the interviews were held with employees working in different fields.
The secondary data, which are data that were already published in the past, were collected from a number of other sources. First, Philips provided access to a range of documented material available on their ‘Grooming SharePoint’. This is a portal for employees to have access to all information concerning current and previous grooming projects. Other internal corporate documents were consulted as well. The internet and academic literature such as books and articles were also used for conducting this research.
4. Analysis and diagnosis
Based on the problem statement, which was derived in section 2.6, this chapter contains the analysis and diagnosis phase which is the third step in the model of van Aken et al. (2007). In this chapter, the current state of the organization will be compared with the theories from the literature. The first section begins by sketching some background information about outsourcing R&D and early supplier involvement. After that the second section continues by describing the PCB design outsourcing process, which answers the first subquestion. Then the third and fourth sections discuss supplier involvement and supplier responsibilities respectively and provide answers to the second and third subquestions. Finally the communication process is discussed in the fifth section to give the answer to the fourth subquestion.
4.1 Background information
This section begins by giving some background information about outsourcing research and development and about early supplier involvement in the first and second subsections respectively. This will done to inform the reader about the reasons why the ESE department increasingly involves suppliers during the development process.
4.1.1 Outsourcing research and development
Page. 27 tasks being performed. One function that is increasingly being targeted for outsourcing is PCB design (Zaccari, 2007). Table 4 shows the many reasons why firms outsource their R&D and the authors who support this.
Advantages of outsourcing R&D Supported by
Reduction of costs Caudy, 2001; Piachaud, 2002; Kumar and Snavely, 2004; Zaccari, 2007
Focus on core competences Piachaud, 2002; Zha and Calantone, 2003 Resource flexibility Piachaud, 2002; Zha and Calantone, 2003 Minimize business risks Caudy, 2001; Piachaud, 2002
Accelerating product development Zaccari, 2007 Building expertise by exposing the internal development staff
to new knowledge, technology and organizational development processes.
Ernst, 2000; Zha and Calantone, 2003
Table 4: Advantages of outsourcing R&D.
As can be seen, literature indicates several advantages for firms that outsource their R&D. However, there are also risks associated with outsourcing. Table 5 shows these risks and the authors who support this.
Risks of outsourcing R&D Supported by
The inability of outsource providers to perform Zaccari, 2007; McIvor et Al., 2009 Loss of skills McIvor et Al., 2009
Miscommunication issues Zaccari, 2007 Organizational change implication McIvor et Al., 2009
Table 5: Risks when outsourcing R&D.
4.1.2 Early Supplier involvement
Rapid technological development, shorter product life cycles, more competition, and increased outsourcing have prompted many firms to involve their suppliers early in their new product development activities (Mikkola & Larsen, 2003). Early supplier involvement (ESI) is generally defined as ‘a form of vertical cooperation in which manufacturers involve suppliers at an early stage in the product development and/or innovation process’ (Bidault et al., 1998). One of the motivations behind ESI is to decrease product development times, reducing product costs, improve quality and also to tap into the suppliers technological capabilities (Mikkola & Larsen, 2003; McIvor & Humphreys, 2003). In appendix V one can read more about the most important benefits of ESI.
4.2 The PCB design outsourcing process
4.2.1 The new product development process
The ESE department calls the NPD process an Integrated Product Development (IPD) process. Because literature calls this IPD process the NPD process, the term NPD process will be consistently used throughout this research. The NPD process of the ESE department looks like the stage-gate-process described by Cooper (1990). Figure 11 shows the NPD process, which is divided into six stages, with each stage providing input for the next stage. The goal of this NPD process framework is to have a formal model for the steps that need to be taken in a NPD process. In appendix VI one can find the whole NPD process, the particular activities for electronics and the deliverables for each stage of the NPD process.
Figure 11: The NPD process of the ESE department.
4.2.2 The PCB design outsourcing process
In this subsection, the PCB design outsourcing process between the ESE department and its suppliers is described. It also discusses where problems occur in this process. Figure 11 shows the six stages of the NPD process, but this figure also shows a summary of the activities during the PCB design outsourcing process. In appendix VII, one can find more detailed information about the whole PCB design outsourcing process and the problems that occur during this process.
Page. 29
4.3 Supplier involvement
This section elaborates on the second subquestion, which is the involvement of suppliers during the PCB design outsourcing process. A questionnaire in appendix VIII was used to get more insight into supplier involvement. The first subsection discusses the level of supplier involvement in the NPD team. The second subsection continues by discussing the current level of supplier involvement of the NPD process.
4.3.1 Supplier involvement in the NPD team
The earlier suppliers are integrated and involved in the NPD process (stage 1 and 2), the better the project outcomes (Handfield and Lawson, 2007). Integrating suppliers in these early stages allows them to bring in their ideas or problems into the NPD team of the ESE department. Although this seems like a common sense approach, the questionnaire shows that there is currently only ‘medium’ participation and involvement of the supplier in the NPD process of the ESE department. As a result, essential information is not shared with each other timely. This also leads to problems being discovered too late, resulting in rework. In contrast to this, literature suggest that by fully incorporating suppliers on project teams, potential problems can be identified early, thus improving the quality of the final product, eliminating rework and reducing costs (Handfield, 1994). It also enhances the information and expertise regarding new ideas and technology (Smith and Reinertsen, 1991). So a better way seems to integrate the supplier fully into the NPD team of the ESE department. This can for instance be done by letting the NPD team of the ESE department collaborate more closely with the NPD team of the supplier. In case of a problem, both teams can help each other find a solution. Ideas and suggestions can also be exchanged between both teams.
From in-depth discussions with employees of the ESE department, it also appears that the ESE department is not enough involved in the design-in-workshops with a selected supplier at the beginning of a new project. Design-in-workshops are held with suppliers and with different disciplines within Philips to discuss the rough concept of a new grooming product at the start of a new project. Disciplines such as mechanical engineering are very much involved during these workshops in contrast to the electronics discipline. Because the involvement of the ESE department is so small, one of the first steps seems to be to get the ESE department fully involved during these workshops. When they are early involved into this process, they can also bring in their ideas and can discuss feasibility issues about the electronic design.
4.3.2 Supplier involvement during the NPD process
Figure 12: Supplier involvement during the NPD process.
According to Handfield et al., (1999), the input of suppliers or their active involvement during the NPD process can be sought at any point in time during the NPD process. Figure 13 shows these integration points. In addition to this, Blenkhorn and Noori (1990) state that ESI (stage 1 and 2) in the NPD process is better and leads to a reduction of development time and increased product quality.
Figure 13: Possible supplier integration points (source: Handfield et al., 1999).
Page. 31
Figure 14: A more detailed supplier integration model (source: Handfield et al., 1999).
The PCB designs of previous projects (see subsection 2.2.2) that were outsourced were ME and HE appliances. From in- depth discussions with employees of these projects it appears that during these projects the suppliers were involved from the beginning until the end of the NPD process. So it seems that the ESE department has tried to use ESI, and they succeeded to successfully involve the supplier at every stage of the NPD process. However, the reason that these projects cost more R&D hours and longer throughput times was because the supplier responsibilities were unclear (see appendix III). The model showed that the ESI approach, which involved the supplier in the project from the beginning until the end, is the best approach in case of more complex items. This model can function as a starting point for the ESE department to determine when to involve their supplier into the NPD process. The ESE department has therefore to split up its projects into two categories (see the red dotted lines in the model) and treat these two categories differently, by involving the supplier earlier or later in the NPD process. The next section will continue by elaborating on the supplier responsibilities in the PCB design outsourcing process.
4.4 Supplier responsibilities
This section elaborates on the third subquestion, which is the responsibility of suppliers during the PCB design outsourcing process. A questionnaire (see appendix VIII) was used to get more insight into this responsibility. The first subsection gives information about a project book which is made during each new project. The second subsection continues by mapping the supplier responsibility into a model. The third and fourth subsections discuss a grey and black box design.
4.4.1 Project book
design are currently not formalized in this project book. Because of these unclear responsibilities, the ESE department and the supplier do not know what to do in case of a problem. If a problem occurs, the ESE department currently always takes full responsibility by jumping into the NPD process to help their supplier, also partly because of the passive attitude of suppliers. In previous projects (see subsection 2.2.2), the ESE department had also helped the supplier create a working concept. However, this had led to many extra hours and costs. In addition to this, Hartley et al., (1997) argued that by shifting responsibility for component design to the supplier, the buyer’s engineers can focus on the buyer’s own core design competencies and assume other tasks, such as coordinating the design activities of the supplier. In that case the ESE department should place the full responsibility for the PCB design at the supplier, in order to free up time for coordinating activities, such as monitoring progress. Dyer and Ouchi (1993) also support this by saying that Japanese companies assign relatively complete subsystems to suppliers and give their suppliers significant design responsibility. It seems therefore important to formalize all responsibilities with respect to the PCB design in the project book when a new project is started. When each party knows its responsibilities they know what to do in case of a problem and how to act during the NPD process.
4.4.2 Mapping supplier responsibility
To get a better understanding of the level of supplier responsibility, the supplier responsibility needs to be mapped in order to determine where the ESE department currently stands and where it can go. To measure the level of supplier responsibility, the model of Handfield and Lawson (2007) is adopted (see Figure 15). Table 6 explains the different stages of this model. In this model, the supplier responsibility ranges from no involvement, to “White box” (the supplier is consulted informally on the project), “Gray box” (there is a formalized joint activity that takes place), or “Black box” (the design is primarily supplier-driven based on the buyer’s performance specifications). This latter form of involvement constitutes the largest amount of responsibility for the supplier.
Integration level Description
White Discussions are held with suppliers about specifications/requirements but the buying company makes all design and specifications decisions.
Grey The buyer and supplier enter into an informal or sometimes a formal joint development effort, which may include information and technology sharing and joint decision making regarding design specifications.
Black The supplier is informed of customer requirements and then is given almost complete responsibility for the purchased item, with only review and concurrence on the purchased item's specifications by the buying company.
Table 6: Level of supplier integration (source: Handfield and Lawson, 2007).
Page. 33 specifications and requirements. By knowing this information we can now place the supplier responsibility into the model of Handfield and Lawson (2007). Figure 15 shows that the current level of supplier responsibility is now in the white box, as indicated with a light blue color. This is because there are sometimes discussions between the ESE department and the supplier but the ESE department finally takes all design and specification decisions for all PCB designs (LE, ME and HE). It seems that the supplier responsibility in this box is quite low.
Figure 15: Supplier responsibility (source: Handfield and Lawson, 2007).
4.4.3 Gray box design
It became clear that the ESE department currently always takes full responsibility, which leads to a lot of extra hours and costs (as was shown in the previous projects, see subsection 2.2.2). In addition to this, Clark and Fuijimoto (1991) argue that by increasing suppliers’ responsibilities for component design has been credited with decreasing buyer’s engineers' hours. Because supplier responsibility in a white box is quite low, the best approach seems to be to start using a gray box design which can function as an intermediate stage to work towards a black box design. Because a black box design approach is probably still a bridge too far for the ESE department as yet. Figure 15 illustrates this path by two arrows that indicate the growth direction from a with box design to a gray box design to finally a black box design. In the case of using a gray box design, suppliers can participate in the product development effort by sharing their tacit knowledge, skills, information, and technology (Xenophon et al., 2007). Suppliers can also join the decision-making regarding design specifications. Figure 15 shows that with a gray box design suppliers are getting more responsibility, also for the PCB design. When the ESE department grows from a white box design to a gray box design, they can accustom the supplier to having more responsibility for the PCB design.
4.4.4 Black box design
detailed component design from the buyer to the suppliers by using black box designs, reduced both development time and costs. A black box design is also desired because it enables suppliers to offer their ‘off the shelf’ solutions to the ESE department (see also a supplier development grooming document in appendix X).
However, black box designs are only possible when the subsystems of the product can be modularized and its interfaces can be standardized (Mikkola and Skjoett-Larsen, 2003). Another premise is that the supplier must be capable enough to use a black box design (and also a gray box design). If not, the buyer firm should either help the supplier or substitute the supplier for a better one. However, most suppliers are still not mature and skilled enough to use black box designs (see appendix X). Substituting the supplier for a better one is no option, since suppliers are not only working on the PCB designs but also on mechanical aspects, housing aspects, etc. A way to overcome the immature character of a supplier according to Handfield et al. (2000) is to run a supplier development program to improve the supplier’s performance. This solution is also stated by the supplier development grooming document in appendix X. Such a program has recently been started on a small scale with one supplier (Supplier 5).
4.5 Communication process
In this section the fourth subquestion will be elaborated on by describing the communication process between the ESE department and their suppliers. A questionnaire (appendix VIII) was used to get more insight into this process. First the social and technical communication instruments that the ESE department uses are discussed in the first and second subsections. Then the communication process between the ESE department and its suppliers is analyzed in the third subsection.
4.5.1 Social communication instruments
The ESE department communicates with its suppliers several times a week. As was revealed by the questionnaire, the most commonly used communication means are email, telephone and telephone conferences (TC’s). The ESE department uses these communication means to communicate with its their suppliers several times a week to keep track of the project, to solve problems, etc. Literature supports this by saying that frequent communication between the buyer and supplier during product development ensures that information is shared on a real-time basis, thus allowing the supplier as well as the buyer to respond more quickly to the inevitable changes that occur during product development (Hartley et al., 1997). In addition to this, Ancona and Caldwell (1990) argue that high levels of communication both within the product team and externally with stakeholders are required for the success of an R&D project. So it can be argued that the weekly communication of the ESE department with its suppliers is a good point.
4.5.2 Technical communication instruments
Page. 35 frustration among engineers of the ESE department, especially when engineers of the ESE department need to check and change the supplier’s CAD drawing of a supplier. In that case they always have to ask the supplier’s engineers to make little changes, which leads to delays and sometimes wrong adaption’s of the CAD drawing. When a CAD drawing is finished, a Gerber file is generated. A Gerber file is a standardized fixed file format to describe the images (solder mask, copper layers, traces, vias, etc.) of the PCB as well as the drilling data. The ESE department and the supplier can both read these Gerber files. However, the problem is that different CAD file formats cannot be read by the ESE department. So it seems that the CAD file format is leading and it is easier to have one common CAD file format.
Another problem is that problems occur with respect to the version management of CAD drawings. There are cases when engineers of the ESE department do not know which is the latest CAD drawing. The ESE department currently uses Altium for its CAD drawings, while suppliers use different CAD programs such as Eagle, Ultiboard, old versions of Altium, etc. A case study into the electronics industry from literature suggest, however, that common, linked information systems are important because this allows the customer and the supplier to easily share information with each other (McIvor and Humphreys, 2009). Literature shows that it is better for the ESE department and its suppliers to use common, linked information systems. The ESE department currently already uses conversion programs to read CAD files from suppliers that use different file formats. However, using the same CAD program would be better. Although it might be an option to prescribe Altium as the standard CAD program to the supplier at the beginning of a new project, this may result in the supplier calculating extra costs. So there is a trade off that needs to be made. The supplier also has to acquire the skills of working with a new CAD program, like Altium. In order to learn this CAD program correctly, it may be possible to run a pilot program. When the supplier finally has enough knowledge and skills to use Altium as their standard CAD program, the ESE department should prescribe this program into the supplier conditions at the start of a new project. Such information can be formalized in the project book (see sub section 4.4.1).
4.5.3 Communication process
It also appears that during the communication process information is wrongly interpreted and some information lacks. This wrong interpretation can, for example, occur because of the cultural differences between Dutch and Chinese engineers. Also the absence of rich communication (face-to-face) can be a reason that wrong interpretations take place. In-depth discussions have shown that incorrect interpretations mostly occur with respect to ME and HE PCB designs and less with LE PCB designs. This is probably because LE PCB designs are simpler and less complex, while ME and HE PCB designs are more complex and require more tacit knowledge. Another issue is that, to a small extent, some information lacks, for instance information about the setup of testing tools, trial run results, availability of components, etc. The framework of Daft and Lengel (1986) is used to see where in the model the current communication process between the ESE department and their suppliers can be located. Figure 16 shows this framework. In appendix IX one can read more about this framework. Wrong interpretations (equivocality) and the absence of information (uncertainty) are two forces that exist in organizations that influence information processing (Daft and Lengel, 1986). These two forces are placed on the vertical side (equivocality) and horizontal side (uncertainty) of this framework. From the questionnaire it appears that during the communication process, equivocality is higher than uncertainty. The blue box (ESE) in the framework gives the current position of the ESE department. This means that it is currently more in cell 1, which says that the communication process has a high equivocality and a low uncertainty. A more preferable state for the ESE department would be to have both low equivocality and low uncertainty, that is to be in cell 3, which is the ideal state. This is illustrated with an arrow in the figure.
Figure 16: Adapted framework of equivocality and uncertainty (source: Daft and Lengel (1986).
Page. 37 information and rich communication like face-to-face communication, for instance during group meetings, is most effective because it transfers information through multiple cues and allows for immediate feedback and personalization. Frequent communication using a rich medium allows the buyer and supplier to develop a common interpretation of information and thus reduces delays because of mistakes and misunderstandings (Daft and Lengel, 1986). Face-to-face which is the richest communication medium is followed by telephone conversations, and then written communication. These latter two communication methods are currently being used. A recommended way is to co-locate a portion of the electronic team of the supplier, by expatriating them for an extended period to work on-site at the ESE department in a so-called global business team. These expatriate teams must manage the ambiguity, conflicts, and challenges that are posed by having multiple and diverse organizations collaborating on a single project (Takeishi, 2001). A case study of Japanese product development practices has also highlighted the benefits of face-to-face buyer-supplier communication during product development. In this study the authors suggest to make use of ‘guest design engineers’ to work side by side with their customer’s engineers at their customer’s facility during product development (Kamath and Liker, 1994; Blenkhorn and Noori, 1990). Ragatz et al. (2007) argue that co-location is only used in case of technically complex items and that co-location is not used from the beginning until the end of the NPD process, but only for specific activities, such as problem solving activities, prototype testing etc.
5. Plan of action and conclusions
This chapter presents a redesign which can be seen as a process towards improvement of the diagnosed problem from chapter 4 by providing concrete solutions (De Leeuw, 2002). Van Aken et al. (2007) call it a plan of action, which is the fourth step in their model. This chapter gives solutions in the form of recommendations based on the analysis and diagnosis phase of the previous chapter. These recommendations relate to subquestions 2, 3 and 4 of the problem statement and the research questions (see section 2.6). The first section gives an introduction of the recommendations of Figure 17. The second section explains these recommendations. The third section discusses the current capability of each supplier with respect to all recommendations. The fourth section concludes with a generalization of all results.
5.1 Summary of the recommendations
The objective of this research was to shorten the throughput time and to lower the costs of the electronic design for grooming products. The research question that was formulated to reach this objective was stated as follows:
• “How should the ESE department improve their PCB design outsourcing process to make a clear outsourcing decision, which leads to a reduction in costs and throughput time?”
The solution that is obtained consists of a series of recommendations. Figure 17 states the redesign which is the solution for the research question above. It must be noted that this redesign is placed on a higher aggregation level than the detailed level of the PCB design outsourcing process (see appendix VII). It is the intention to implement the solutions from this redesign trough the whole PCB design outsourcing process. Figure 17 is split into two categories which are ‘PCB design LE grooming appliances’ and ‘PCB design ME/HE grooming appliances’. This split-up is made because for some activities each category needs to be treated in a different way. For other activities both categories can be treated the same. The reason why the PCB design of ME and HE grooming appliances are taking together is because their complexity is quite similar and because they have almost the same functions (see section 1.3). The PCB design of LE grooming appliances are less complex and has less functions. It is therefore treated as another category. Figure 17 has three steps and their associated activities.
