Supplier Assessment at Cordis

`

Supplier Assessment at Cordis

Monitoring and Improving Supplier Performance

Annemarie Doedens

Roden, December 2003

Supplier Assessment at Cordis

Monitoring and Improving Supplier Performance

Date: December 2003

Place: Roden, The Netherlands Author: Annemarie Doedens

First supervisor: Prof. Dr. D.J. Kamann

Second supervisor: Drs. D.F.F.R. Maccow

Internal supervisor: Mr. Drs. S.D. Bloemsma

Faculty of Management and Organisation

University of Groningen

Management summary

The subject addressed in this research is that of supplier assessment. In the current situation suppliers at Cordis are evaluated every quarter by way of a supplier rating. The performance- indicators used are product quality and on-time delivery. However, it is argued that for a deeper understanding of supplier performance it will be necessary to improve the model. The research question has been formulated as follows:

What should a Cordis specific Supplier Assessment Model look like, with the help of which supplier performance can be monitored and improved?

To answer this question, four sub-questions have been formulated. The first sub-question deals with the type of performance measurement method most suitable for Cordis. It was not conclusive whether the supplier rating method currently used is in actual fact the most appropriate method for Cordis. Hence, different types of measurement methods mentioned in literature have been studied. The most suitable method was determined by examining the purpose for which it is intended, the buying situation at hand, and the supply risk considered.

The main purpose Cordis wants to use the supplier assessment system for is described in the problem statement: to monitor and improve supplier performance. Supplier rating was found to be the most suitable performance measurement method for the situation at hand, provided that the assessment model reflects both objective (hard) performance-indicators and

subjective (soft) performance-indicators.

A group decision-making model was used to find out which performance-indicators should be included in the assessment model for Cordis. The final choice was made by the clients of the process, with the restriction that the criteria should be measurable. This was the reason why a definite answer to the question “which criteria should be included in the Supplier Assessment Model for Cordis” could only be answered after elaborating on the third sub-question “how can we measure the assessment criteria determined in sub-question 2”. Moreover, the discussion about the importance of every criterion was encouraged in the process of making the criteria operational, which sometimes led to new insights.

The definite model consists of five categories, derived from the categories distinguished by Kannan and Tan (2002). Every category was assigned a weight, based on the perceived importance of the category in question. The categories are:

1. Quality (35%);

2. Delivery and Service (30%);

3. Responsiveness and Development (15%);

4. Information sharing (15%); and 5. Price (5%).

Every category consists of three criteria, except for Responsiveness and Development which consists of four. For the different criteria a certain score can be obtained, which should add up to 100 per category. While Quality, Delivery and Service and Price include almost only criteria that are measured objectively, the other two categories consist for a large part of subjective criteria. The subjective criteria are measured with the help of a “checklist” that defines the criterion by enumerating several issues related to it. By reflecting on these issues the criterion can be measured in a relatively objective manner.

By introducing a Supplier Assessment Model, supplier performance at Cordis can be

monitored. Based on the results it should be decided whether improvements in a certain area are desired. The model clearly shows in which area possible improvements are needed.

Since it is a flexible system it should be adjusted to changing circumstances. Moreover,

attention should be paid that not too much emphasis is put on an incidental low score; more

important is to look at the performance of a supplier in the long-term, and to detect patterns of

change over regular intervals of time. It is important that the performance assessment is not

only used to penalize or reward suppliers; rather it should help to make things visible, to show

suppliers where they are and the direction they should follow.

Acknowledgements

The thesis lying in front of you has been written to finalize my study in Business Studies at the faculty of Management and Organization of the University of Groningen. It is a research project about supplier assessment, carried out in order of Cordis Europe N.V. in Roden, The Netherlands. Writing this thesis required a lot of involvement and perseverance. It also required the help of many people, and I would like to take the opportunity to show my appreciation to those who helped me to bring this research to a successful ending.

First and foremost, my gratitude goes to Suzanne Bloemsma, my internal supervisor at Cordis, who gave me the opportunity to carry out this research and whose optimism and enthusiasm during the entire research project were very contagious. Furthermore, special thanks goes to Harry Huizeling, who was always there to answer my questions, and of course to all the others who contributed to this research and/or to the nice time I had at Cordis.

Next, I would like to thank my supervisors of the Faculty of Management and Organisation of the University of Groningen. My first supervisor, Prof. Dr. D.J. Kamann, for his helpful comments and constructive criticism, and a similar appreciation goes to my second supervisor Drs. D.F.F.R. Maccow, who largely contributed with advice on the structure and methodology. I want to thank both for the time they invested in my research project.

Last, but certainly not least, I owe much gratitude to my friends and family, who are always there with help and good advice. My special thanks goes out to my parents and my sisters, who supported me all the way through my studies, in good times and bad times.

With this thesis, also my student time has come to an end. It means stepping into the world of the working people, which is certainly a new challenge. I will always remember my time in Groningen with much pleasure.

Annemarie Doedens

Roden, December 2003

Table of Contents

Chapter 1 The Organisation ... 1

1.1 C

... 1

1.2 C

J

& J

... 2

1.3 C

... 2

1.4 C

... 3

1.5 S

Q

A

S

M

... 4

Chapter 2 Research Outline ... 5

2.1 I

... 5

2.2 P

... 6

2.2.1 Reasons for research: an indication of existing bottlenecks... 6

2.2.2 An overview of the current situation ... 7

2.2.3 Description of the existing problems and formulation of the assignment... 9

2.2.4 Rules and regulations with regard to Supplier Assessment... 12

2.3 P

... 14

2.3.1 Research question and preconditions ... 14

2.3.2 Conceptual model ... 15

2.3.3 Sub-questions ... 16

2.4 M

... 17

2.4.1 Type of research ... 17

2.4.2 Data collection and data sources ... 17

2.4.3 Research planning ... 19

2.4.4 DOV vs. Process Excellence... 19

Chapter 3 Performance Measurement Methods ...23

3.1 I

... 23

3.2 S

... 23

3.2.1 Reasons for using supplier performance-indicators ... 24

3.2.2 Supplier evaluation vs. Supplier rating... 26

3.3 D

... 27

3.4 S

... 30

Chapter 4 Supplier Assessment Criteria for Cordis ...34

4.1 I

... 34

4.2 A

-

(

) ... 34

4.3 S

C

... 39

4.3.1 Identification of the active participants to be involved in the decision-making ... 39

4.3.2 Identification of the significant criteria involved in the decision-making ... 40

4.3.3 Ranking of the alternatives... 41

4.3.4 Obtaining the consensus ranking... 44

Chapter 5 Making the Criteria Operational...45

5.1 I

... 45

5.2 D

... 45

5.2.1 Quality ... 46

5.2.2 Delivery and service ... 48

5.2.3 Responsiveness and Development... 50

5.2.4 Information Sharing... 52

5.2.5 Purchase price... 54

5.3 T

D

S

A

M

C

... 56

5.4 A

... 57

5.4.1 Relational vs. operational elements of service performance ... 57

5.4.2 Service level agreements ... 58

5.4.3 Concluding remark ... 60

Chapter 6 Making the Supplier Assessment Model Operational ...61

6.1 T

O

S

M

... 61

6.2 I

:

... 62

6.3 C

-

... 64

Chapter 7 Conclusions and Recommendations ...65

7.1 C

... 65

7.2 R

... 66

7.2.1 Recommendations... 66

7.2.2 Future research ... 67

Bibliography...70

1

Chapter 1 The Organisation

For more than 40 years, Cordis Corporation, a Johnson & Johnson company, has pioneered less invasive medical devices for vascular disease. Technological innovation and a deep understanding of the medical marketplace and the needs of patients have made Cordis the world’s leading developer and manufacturer of products for interventional medicine. Other important actors in this market are Boston Scientific and Guidant.

Besides the office in Roden, for which this research project is carried out, Cordis has offices in the United States, in Mexico and in Puerto Rico. The corporate headquarters are

established in Miami Lakes.

In this chapter the organisation will be introduced. We will start with an overview of the company history, after which a description of the core business and the company structure follows. In the last paragraph the Supplier Quality Assurance department and the Supplier Management will be scrutinized.

1.1 Company history

In 1959 Dr. Wm. P. Murphy founded the Cordis Corporation in Miami, Florida, as a medical device corporation and rapidly gains recognition for being a pioneer in innovative devices and products for interventional vascular medicine and electro-physiology. The first product Cordis developed was an injector, a device with which contrast fluid could be injected under pressure in the vascular system. At the same time a pacemaker was developed, which was implanted in the human body for the first time in 1962. At this time, doctors still made their own

catheters. The Cordis Corporation was the first organisation to market pre-shaped and sterilised cardiological catheters on the American and Canadian market.

In 1966 a small business enterprise in Veenendaal got permission to represent Cordis in Europe. They decided to establish the European plant in Roden, and in 1970 the first products were produced and launched on the market. The main products Cordis

manufactured were pacemakers. The fundamental research was carried out in Miami, but in Roden the need for technical expertise grew. The strict legislation in the United States and the development of the European market made it necessary to start the research and development of new products in Roden.

The pacemaker production was sold in 1987. Cordis enters the interventional marketplace with a full line of PTCA (percutaneous transluminal coronary angioplasty) guiding catheters.

The technology becomes the industry standard. The research and development department in Roden was granted the responsibility for the whole design and development of diagnostic catheters. Cordis only made catheters for the heart (cardiology), but the growing need for catheters for other parts of the human vascular system (radiology) did Cordis decide to focus on radiology as well.

In 1990 Cordis introduces the first PTCA balloon utilizing nylon balloon technology. This material also becomes the industry standard. By inflating the balloon on the place of the stricture in a blood vessel, the stricture could be improved. The researchers in Miami

developed a radiological balloon-dilatation-catheter. The production in Roden started in 1992.

A year later Roden was the first with the production of cardiological balloon-catheters. For both successful production lines Roden was given the responsibility for further development, and as a result the product-offer expanded enormously.

The development of new products has a high priority. In 1990 Johnson & Johnson

Interventional Systems Company receives FDA clearance to market the so-called stent, which

is cited by physicians as a “breakthrough in interventional medicine”. Stents are slotted steel

tubes developed for strictures in the vessels, which are often the result of a disease of the

vascular wall. When the stricture is pushed open by a balloon-catheter, a new stricture often

2 appears on the same place. Stents are developed to prevent these vessels to slip again. A stent is crimped on the balloon-catheter and during the procedure it is permanently left behind in the blood vessel.

Another important development was the Cypher™ stent, a drug-eluting stent to evaluate the risk of in-stent restenosis. This stent significantly reduces late lumen loss and reduced the incidence of restenosis to zero after placement in patients with coronary artery disease.

In 2002 Cordis receives approval in Europe for the Cypher™ Sirolimus-eluting stent, the first approved drug-eluting stent, signalling a new era in the treatment of cardiovascular disease.

In 2003 Cordis received FDA approval to market its Cypher™ stent in the United States, making it the first drug/device combination for the treatment of restenosis.

The next few years Cordis will aim at continuous cooperation to attain the best results. The objective of Cordis is to become number one in the area of treatment and diagnosis of heart- and vascular diseases. In Roden the fast development of new products for heart-and vascular diseases at the lowest cost is at the centre.

1.2 Cordis and Johnson & Johnson

On the 23rd of February 1996 Cordis Corporation merges with Johnson & Johnson

Interventional Systems Co. to form Cordis Corporation, a Johnson & Johnson company with approximately 3,500 employees worldwide. Johnson & Johnson is globally an active player in the healthcare industry. The Johnson & Johnson head office is situated in New Brunswick, New Jersey. In total they own 188 companies in more than 52 countries. The products are sold in more than 175 countries. Well over 100,000 employees work for Johnson & Johnson in the following three branches:

1. Consumer products 2. Pharmaceutical market 3. Professional market

Johnson & Johnson is very concerned about renewal and invests about 10% of its’ turnover in Research and Development. The success of Johnson & Johnson is owed to the company philosophy, which has been expressed in the company Credo. The Credo is the basic philosophy of Johnson & Johnson and contains all norms and values which Johnson &

Johnson employees have to respect. Clients occupy the first place, followed by employees, the community and finally the shareholders. The most important lines of policy are renewal, the delivery of quality as a motive, inspiring leadership, good support for employees, and customer orientation.

1.3 Core business

The core businesses of Cordis are:

1. The development of new products. The department product development is

occupied with the development of new products or improvements of existing products.

2. The production of new products. The products Cordis manufacturers are produced under strict requirements. To meet these requirements the organisation possesses 9000 m² clean room. In these dust-reduced areas production can take place under controlled circumstances. Unique for Cordis Roden is the presence of an own 100%

Ethylene Oxide steriliser, by which the products can be sterilised in-house.

3. The distribution of products. Supply Chain Management stores all Cordis-products that are produced elsewhere before further distribution in Europe. Roden ships the end products to Johnson & Johnson and Cordis companies, who distribute the products to the hospitals.

Furthermore, Cordis Roden is occupied with legislation with regard to medical devices. The department of Quality Compliance Services makes sure that the legislation in countries where Cordis-products are sold is met. Other important activities of Cordis Roden are quality

assurance, which will be discussed in paragraph 1.5, and customer support.

3

1.4 Company structure

Cordis has two business units: Cordis Cardiology for cardiovascular disease management and Cordis Endovascular for the treatment of peripheral vascular and obstructive diseases.

Cordis Roden is managed by site management. This means management is responsible for all general and operational issues related to the plant in Roden. The site management

consists of the management team led by the chairman. The directors of the main departments all form part of the management team. Every department has a few sub departments led by department managers. The sub departments are divided into groups. In figure 1 the company structure is visualised. The Quality Assurance department is taken as an example, because this is the department for which this research is carried out.

Figure 1: The company structure at Cordis

Cordis is made up of the following main departments:

1. Operations (production) 2. Engineering

3. Supplier Management (purchasing)

4. Supply Chain Management (distribution centre) 5. Safety and Environment

6. Regulatory Affairs 7. Quality Assurance 8. Information Management 9. Human Resources 10. Finance

11. Materials Management

12. New Product Development

4

1.5 Supplier Quality Assurance and Supplier Management

American and European legislation set high requirements with regard to the quality of medical devices. The Quality Assurance department has to secure that the Cordis Quality System is in compliance with the legislation. The department is divided in the following four groups:

•

QA Operations PTA (percutaneous transluminal angioplasty)

•

QA Operations PTCA (percutaneous transluminal coronary angioplasty)

•

QA Logistics, Purchasing and CAPA (corrective and preventive action)

•

Complaint Analysis

Together these four groups make sure that Cordis works according to the generally received opinion. Cordis’ main responsibilities are the physicians, patients, family, and all the others that use their products.

This research is carried out for the department Supplier Quality Assurance (SQA), which is part of QA Logistics and Purchasing. SQA contributes to the assurance of purchasing quality, which is important for both physical items and services, for each can affect the quality of the overall product shipped to the customer. The three main activities of SQA are:

•

Qualification of suppliers and materials, based on specifications;

•

Monitoring and Corrective Action Management;

•

Evaluation, development and Certification.

By carrying out these three activities the supply of qualitatively good products and services should be assured. Figure 2 represents the supplier quality assurance process. It is the evaluation process that will be scrutinized in this thesis.

Figure 2: The supplier quality assurance process

Supplier evaluation is also a task of the Supplier Management department. Whereas SQA is in the first place concerned about the quality of the materials and services delivered, Supplier Management is more concerned with the purchasing-process, outlined in figure 3.

Figure 3: The purchasing process It is the task of Supplier Management to:

Select suppliers;

Conclude contracts with suppliers;

Conclude confidential agreements with suppliers;

Make price-appointments with suppliers;

Control the purchasing process.

The Supplier Management department distinguishes two areas of attention. Firstly, the components-area, which addresses materials and services that are used in the end products.

It concerns the strategic and cost side of the purchasing process and is carried out in close cooperation with the Materials Management department. Secondly, the non-inventory &

capital area, which addresses the processing of internal purchasing-requests for materials

and services that are not included in the end product.

5

Chapter 2 Research Outline 2.1 Introduction

In this chapter the framework of the research will be outlined. Since the nature of the research is problem solving we will commence with a diagnostic preliminary research (paragraph 2.2), in which the diagnostic problem statement is described. The diagnostic problem statement reproduces the management problem: a skilful representation and interpretation of the problem tangle (De Leeuw, 1996: p.201). First of all a short description will be given of the reasons for research, after which the existing (problem) situation will be outlined. The reasons for research together with the analysis of the problems ascertained will be translated into the main problem statement of this research.

By means of the main problem statement, outlined in paragraph 2.3, an attempt will be made to find an answer to the diagnostic problem statement. The main problem statement consists of three elements: the research question, the research objective and the preconditions in relation to the research. The research question has been divided into four sub-questions, which are introduced and discussed in paragraph 2.3.3.

The research approach will be determined on the basis of the problem statement. Special attention will be paid to the data collection methods and the data sources that will be used during this research project.

This chapter is structured as follows:

§ 2.2 Problem analysis

•

Reasons for research: and indication of the existing bottlenecks

•

Overview of the current situation

•

Description of the existing problems and formulation of the assignment

•

Rules and regulations with regard to supplier assessment

§ 2.3 Problem statement

•

Research question and preconditions

•

Conceptual model

•

Sub questions

§ 2.4 Methodology

•

Type of research

•

Data collection and data sources

•

Research planning

•

DOV vs. Process Excellence

In the consecutive chapters the four sub-questions will be elaborated. The last chapter will

delineate the conclusions and recommendations with regard to this research, as well as

suggestions for further research.

6

2.2 Problem analysis

The reasons for research, as described in paragraph 2.2.1, illustrate an indication of the existing bottlenecks as a provisional idea. In order to formulate a thorough problem statement it is necessary to know what the problem exactly amounts to. With the help of a problem analysis the problem tangle can be described (De Leeuw, 1996: p.183). In this manner a deep understanding is created of the existing problems and the persons in the organisation who are concerned about this. In addition, a problem analysis will also contribute to the relevance of this research project.

In paragraph 2.2.2 an overview of the current situation is presented, followed by a description of the existing problems, and the formulation of the assignment. The paragraph will be concluded with the rules and regulations with regard to supplier assessment.

2.2.1 Reasons for research: an indication of existing bottlenecks

The relationship between a company and its supplier is supported by means of supplier selection, audits and periodical supplier assessment. A win-win relationship will increase the power of both the company and the supplier. At the moment Cordis routinely evaluates its suppliers by means of supplier audits and a supplier rating which is carried out every quarter.

This method is used within all Cordis franchises, i.e. Cordis Roden, Miami, Warren, Mexico and Puerto Rico. The assessment of suppliers is based on information in the ERP-system.

The anxiety that exists about the current assessment method actuated this research project.

In first the instance three bottlenecks were mentioned in this context, which will be shortly discussed.

First of all, the input-data is not always reliable, which obviously has repercussions on the outcomes. For example, quality is determined on the basis of the number of supplier-related MRR’s completed and signed off during a quarter. MRR’s are Material Re-assessment Reports and are they refer to an entire lot of material. An MRR has to be completed when materials do not conform to the specifications. Non-conformities can be discovered by the Receiving Inspection department, where delivered materials enter the company, or during the production process. Completed MRR’s have to be checked and signed by several persons.

Every so often an MRR is checked and signed after the period (quarter) in which it occurred.

As a result the MRR under consideration is not included in the supplier rating for that specific quarter, which makes the quality rating, and thus the total rating, unreliable.

Furthermore, supplier-related rejection can have a negative influence on the reliability of the supplier rating. Rejection refers to a specific “piece” of material instead of an entire lot, and is discovered during the production process. Supplier-related rejection is not taken along in the quality rating. Because of this, it suspected that the actual quality of the materials from a specific supplier is lower than what is demonstrated by the quality rating, due to the high rejection-rate throughout the production process.

Secondly, the current assessment method only evaluates suppliers on two aspects, namely product quality and on-time delivery. The surmise has been raised that besides these “hard”

performance-indicators also more “soft” performance-indicators, like for example the communication with the supplier, or the feeling one has about a certain supplier, should be assessed to get the right impression of a supplier’s performances.

The third bottleneck is that at the moment only suppliers of materials are being assessed,

whereas there are also several service suppliers who make a contribution to the end product

and thus should be formally included in the supplier rating as well. The question is whether

service suppliers can be be assessed on the same criteria used for the assessment of

material suppliers, or whether a separate set of criteria should be developed.

7 2.2.2 An overview of the current situation

Prior to the analysis of the problems suspected, the current situation with respect to supplier assessment will be discussed. Insight in the current situation has been acquired by means of semi-structured interviews with the persons involved in the supplier rating process and by investigating the available documents with regard to supplier assessment. First we will have a look at the suppliers that are included in the supplier rating, after which the supplier rating procedure will be described.

The suppliers included in the supplier rating

Of the 1272 suppliers Cordis deals with, the number of suppliers that are being routinely assessed fluctuates between 25 and 35 suppliers every quarter. It is explained in Cordis Franchise Method 11-012 which suppliers need to be included in the supplier rating. The method applies to production and service suppliers of all purchased or otherwise received product and services that relate to the Cordis Quality System worldwide. It is stated that only qualified suppliers should be included in the supplier rating, and for that reason we will first describe the supplier qualification process.

A special document should be completed for production and service suppliers who deliver a product or service which is production-related. This document is divided into 3 categories:

1. Suppliers of components, products, and materials 2. Service suppliers

3. Consultants

Each category contains a number of statements about the type of supplier under consideration. If the answer to one of these statements is “yes”, the supplier should be qualified, because the purchase will have a direct effect upon the quality of the products, the labelling, the manufacturing environment, or the quality system of Cordis”. If none of the statements apply, the supplier does not need to be qualified.

Qualified suppliers have to meet a number of criteria, which are specified in Cordis Franchise Method 11-012. Once they meet with these criteria they are being added to the “Approved Supplier Database” as qualified suppliers. Once a supplier is qualified, a multidisciplinary team will classify the supplier into a category based on the guiding principles outlined in the

“Classification Guidelines”. According to these guidelines suppliers may be classified as Category 1, Category 2, or Category 3 suppliers. The difference between these suppliers lies in the criticality of the products they supply. Category 1 suppliers deliver the most critical products, i.e. products that are being used inside the human body, for example needles, stents, pouches, and guide wire. Category 2 suppliers are less critical and deliver for example stiffing wire, ID band, tubes and instructions for use. Category 3 suppliers are the least critical and deliver cleaning compound, distilled water, crimp tubes, packaging products, ethanol, and ethylene oxide.

Qualified production and service suppliers are audited periodically. The purpose of the audit is to determine if significant changes have occurred to the quality system since the last audit, update the supplier profile and verify effectiveness of corrective actions to observations from prior assessments. The frequency with which suppliers are being audited depends on the category they are in. Category 1 suppliers are audited most often (no less than every 8 quarters) and Category 3 suppliers are audited the least (no less then every 16 quarters).

The performance of each qualified supplier that has submitted three or more lots within a

specific quarter will be included in the supplier rating at the end of that quarter. At the

moment the qualified supplier list contains 147 active suppliers. If we look at the third quarter

of 2003 we see that only 29 suppliers have delivered three or more lots. Subsequently, only

these 29 suppliers were included in the supplier rating. In the first quarter of 2003 the supplier

rating consisted of 26 suppliers, and in the second quarter the rating included 33 suppliers.

8 The supplier rating process

To create an unambiguous picture of the process a so-called SIPOC will be drawn up. By means of a SIPOC, acronym for Suppliers, Inputs, Process, Outputs, and Customers, the process under investigation can be laid down on a high aggregation level. Quality is judged on the basis of the process output, and can be brought to a higher level by improving the input and the process variables. In order to draw up a SIPOC, first of all the process should be clearly defined. The supplier rating process at Cordis has been roughly described in the Cordis Franchise method. In view of the fact that this is a very limited reproduction of the actual process, which does not lead to a thorough understanding of supplier rating at Cordis, a detailed process description is elaborated and will be described next.

For a start, after materials have been delivered and Receiving Goods has entered the exact delivery date into the ERP-system (BPCS), they are examined by the Receiving Inspection (R.I.) department. Materials Management verifies at the end of each quarter the on-time delivery (OTD) of suppliers. On-time delivery, for every supplier, is calculated with the help of the following formula:

If materials do not comply with the specifications a Material Re-assessment Report (MRR) is completed. Non-conformities can be ascertained by Receiving Inspection, but also during the production process. An MRR is assessed after which a so-called reason-code is assigned to it. All cases of supplier-related non-conformities are assigned a code 80. Only these supplier- related MRR’s are taken along in the supplier rating process.

The Supplier Quality Assurance department (SQA) calculates the Quality Rating (QR) by dividing the number of accepted lots per supplier by the total number of lots the supplier delivered during that quarter. The number of accepted lots is the total number of lots exclusive of the supplier-related MRR’s.

On the basis of the OTD and the QR the total rating can be determined. To calculate this total rating a weight is assigned to both criteria, which leads to the following formula:

The total rating reproduces a percentage on the basis of which a judgement is passed about the performance of a supplier. The Supplier Quality Assurance department and the Supplier Management department verify the outcomes and send them to the supplier. In case of a very low total rating they will contact the supplier in question. When a supplier, during four

quarters, scores less than 95% on the assessment, he is formally evaluated. An important restriction of the process is that only suppliers who deliver three lots or more are being considered in the supplier rating. The situation, as described above, can be clarified with the help of a process-flow diagram (appendix 1).

Now that the supplier rating process has been described in detail, it is possible to create a SIPOC (figure 4). Besides an unambiguous picture of the process, the SIPOC also reproduces very clearly which person in the organization is responsible for the information (input) required. On the basis of this information it is determined which persons should be interviewed to describe the current (problem) situation.

OTD = Purchased item releases delivered on time x 100%

Total purchased item releases due

QR = (lots accepted / lots received) x 100%

Total Rating = 0.80 (QR) + 0.20 (OTD)

9 Figure 4: The SIPOC of supplier assessment

2.2.3 Description of the existing problems and formulation of the assignment In the so-called Voice-Of-the-Customer (VOC) the problem tangle is outlined. The problem tangle, or in terms of Ackoff the “mess”, concerns different but connected problems, perceived by different persons in the organization. These persons can be defined as the “problem- observants”. Problem observants are people experiencing a problem, or in other words “those who are (more or less) concerned about how they think a part of the world is and how it should be” (De Leeuw, 1996: p.175). The term “problem-observant” is more extensive than the term “client”. Also those whose problem solving you do not immediately pursue can perceive a problem. Besides the Supplier Quality Assurance department and the Supplier Management department, who both consider the current method of supplier assessment as a reason for anxiety, also the Materials Management department can be considered as a relevant problem-observant. Materials Management is not a client of the process, but they are in contact with suppliers on a daily basis and therefore very concerned about the topic.

The client is the person perceiving the problem you want to solve particularly. Since this research aims to serve both the Supplier Quality Assurance department and the Supplier Management department we speak of a client-system, which is defined as “the sub-system of problem-observants whose interests you primarily want to serve” (De Leeuw, 1996: p.177).

In addition to the problem-observants and the client-system, we can distinguish the formal

principal of the research project. The formal principal is the person who defined the problem

and who, in most of the cases, also finances the project and gives the official permission for

the research project. Principal of this research project is the Supplier Quality Assurance

department.

10 The complete, worked out Voice-Of-the-Customer in which the problems perceived by the problem-observants are outlined can be found in appendix 2. The problems were revealed by way of interviews, and have been translated into demands regarding the design of the new Supplier Assessment Model. These demands have been summarized in the Critical-To- Quality (CTQ) tree (figure 5). The CTQ-tree helps to move from high-level requirements to detailed requirements, and it ensures that all aspects of the need with regard to the design of the supplier assessment model are identified. The overview has been divided into four main- groups, also called “drivers”, and the forthcoming “critical success-factors”. Both the Voice-Of- the-Customer and the Critical-To-Quality-tree are Process Excellence techniques (see paragraph 2.4.4).

Figure 5: Critical-To-Quality Tree

11 The coloured lines surrounding the critical success factors demonstrate the research-priority given to the different factors. This distinction is drawn up together with the clients of the process.

•

Text outlined in red: critical aspects of supplier evaluation which require further research.

•

Text outlined in orange: needs to be discussed with the departments concerned. No further research is needed at this moment.

•

Text outlined in blue: at the moment investigated in other projects It has been decided to aim the research in the first place at finding adequate and

representative criteria in order to get a good insight in supplier performance. Secondly, we will investigate how these criteria can be measured and weighed in order to show a trend. The rules and regulations with regard to supplier assessment will be shortly discussed in paragraph 2.2.4. An elaboration of the rules-and regulations can be found in appendix 3.

It should be noted that not all observations from the Voice-Of-the-Customer could be encountered in the Critical-To-Quality Tree in exactly the same terms. We will review the observations that are not included.

Firstly, both Supplier Management and Supplier Quality Assurance expressed their concern about the group of suppliers taken along in the supplier rating. This concern specifically relates to the group of suppliers that should be formally included in the supplier rating. The driver “compliance with legal standards” covers this observation. As was explained in the previous paragraph, the suppliers currently included in the supplier rating are all qualified material suppliers that have submitted three or more lots within a specific quarter. Only material suppliers are evaluated; service suppliers are not taken along in the current supplier rating, which particularly worries the Supplier Quality Assurance department. By looking at the rules and regulations it will become clear which suppliers should be formally assessed. We will return to this subject in the next paragraph.

The question whether certain non-conformities should outweigh others is another observation that cannot be found again in the CTQ-Tree. Although it is a very interesting observation, it will be considered outside the scope of this research project. In the first place, it would require a stocktaking of all possible supplier-related non-conformities. A classification method should be put into place with the help of which a score can be awarded, which is based on the importance of the type of non-conformity. Clearly, this can be considered as a research project in itself. It would be too time-consuming to include this observation in this research. It will be noted as a suggestion for further research.

Furthermore, the risk with which Cordis has to deal by throwing in their lot with a supplier was mentioned as a concern. This concern relates to the overall objective of performance

measurement, to serve as a building block for the management control system. Performance- indicators can be described as “qualitative financial or non-financial indicators which

“measure” activities or developments which are of crucial importance to the organisation and in a way function as “informers” which give timely signals and make adjustments possible”

(Jorissen, 1994: p.16). By means of performance measurement Cordis can control supplier

performance and hold suppliers responsible for their functioning. Moreover, performance-

indicators make it possible to call for improvements if necessary. Through monitoring and

improving supplier performance, Cordis can reduce the risk to which the organisation is

exposed.

12 2.2.4 Rules and regulations with regard to Supplier Assessment

Rules and regulations should make clear which suppliers should be formally assessed and whether there are certain standards Cordis has to meet as to how suppliers should be assessed. The external rules and regulations Cordis has to comply with are the FDA (QSR) and the ISO. These standards have been translated into internal standards outlined in Cordis Franchise Method 11-012. The content of CFM 11-012 with regard to supplier rating has been discussed in paragraph 2.2.2. An enunciation of the rules and regulations can be found in appendix 3. In figure 6 a recapitulation of the research results has been portrayed.

Figure 6: Rules and regulations

Both the ISO and the FDA clearly state that suppliers should be evaluated on the basis of their ability to meet specified requirements, however, no general rules in connection with supplier assessment have been laid down. Every organisation should establish its own evaluation method and its own assessment criteria.

With regard to service suppliers the QSR states that services should be treated in basically the same manner as physical components when they affect the quality of the finished device.

The thought and control processes are similar whether one is working with services or with

physical product. Services may include: design activities, various product verification and

validation activities, sterilization, routine maintenance, and calibration of equipment. The

combination of both services and physical components determines the quality of the finished

device. Manufacturers shall establish and maintain the requirements that should be met by

service suppliers that perform a service for them.

13 In the previous paragraph concern was expressed about the group of suppliers currently taken along in the supplier rating. After examination of the rules and regulations, the following can be said about this: the performance of each supplier affecting the quality of the finished device, whether it concerns a physical component or a service, should be evaluated by way of a supplier rating. In other words, all qualified material and service suppliers need to be

included in the supplier rating. As opposed to the current supplier rating, not only suppliers who deliver three or more lots should be evaluated. It happens that a supplier delivers materials every quarter, but no more than 2 lots, with the result that the supplier in question is never taken along in the supplier rating.

In addition, external standards state that the same approach should be employed for service suppliers as for suppliers of physical components, and that the thought and control processes are similar whether one is working with services or with physical product. To assess service suppliers requires that service level agreements are set up between the purchasing

organisation and its suppliers. Service level agreements provide information with regard to performance measurement and deal with monitoring and measuring service performance.

However, at this moment Cordis is still in the process of developing service level agreements

with all service suppliers, and for that reason it has been decided to set up a framework for

the assessment of service suppliers. This framework has been delineated in chapter 5.

14

2.3 Problem statement

The problem statement can be defined with the help of the CTQ-tree (figure 5). The success factors in the red textboxes of the CTQ-tree are those aspects that will be addressed in this research project. The problem statement, which consists of the objective of the research, the research question and the preconditions, has been formulated in agreement with the desires of the clients of the process. It describes “what we want to know and why”, or in a more managerial variant, “what we want to produce and what for”. It constitutes on the one hand the link with the specific problem in actual practice and on the other hand it is a link with the theory, which can be considered as a thinking-tool to get a grip on reality.

2.3.1 Research question and preconditions

The purpose of this research is to design a supplier assessment model for the department SQA of Cordis Europe B.V. with which supplier performance can be monitored and improved.

The new supplier assessment method should give a better insight in the behaviour of suppliers in order to be able to improve their performance by means of structural corrective action, which should lead to a better and stronger relationship between buyer and supplier. By means of continuous performance measurement, deviations of the norms can be quickly determined, and as a result of that intervention can take place in an early stage.

Research question

Preconditions

The preconditions relate to the limitations with regard to the method that will be followed and to the restrictions to make the research feasible (De Leeuw, 1996: p.90). Since this research project should result in a Cordis specific supplier assessment model, we will have to look at the setting in which the model is developed. The current situation at Cordis serves as the starting point for this research project, which means the model is restricted in the first place by the organisation. This also means that the people who have to work with the model will be involved in the development. More specifically, the preconditions for successfully ending the research are:

•

The research has to comply with the requirements of the research organization and the list of demands presented by the faculty of Business Studies of the University of Groningen;

•

The research has to be finished within a period of 6 months;

•

The improved supplier assessment system should comply with internal and external rules and regulations;

•

Successful cooperation between researcher and supervisors is essential;

•

The cooperation of those persons whose assistance is indispensable for the research;

•

The answers to the sub-questions should be relevant and connected to the existing problems.

What should a Cordis specific Supplier Assessment Model look like, with the help of which

supplier performance can be monitored and improved?

15 Relevance, makeability and effectiveness of the problem statement

The problem statement should meet the general requirements of a good problem statement:

relevance, makeability and effectiveness (De Leeuw, 1996: p.89). Relevance is determined by careful examination what the true value will be for the client when the research is finished.

In this case the relevance of the research project is relatively straightforward: it will provide a better insight in supplier performance. By means of this insight performance can be improved, which will be beneficial for the organisation.

The makeability is, in brief, the question whether you are capable to produce a reliable product within the preconditions. Looking at the preconditions one can conclude that no unrealistic requirements are written down and thus it should be possible to fulfil this qualification. And the Effectiveness, after all, is the question whether there is a reasonable relation between the research costs and the benefits. Relevance proves this: the benefits are considerable and will almost certainly outweigh the costs.

2.3.2 Conceptual model

Conceptual models can be described as pragmatic instruments to see and order reality (De Leeuw, 1996: p.140). The approach towards supplier performance measurement and evaluation followed in this thesis will be considered as a conceptual model. This model is based upon the assumption that people behave in a rational way. As Botter (1981) states:

“Rational behaviour requires that the complex reality is replaced with a model that is simple enough to work with. The choice of what parts, items or elements and relations should be included in a model and what not, depends on the objectives of the researcher” (Van Weele, 1984, p.217). Our objective is to use the conceptual model to outline the structure of this research project. The conceptual model therefore includes the fundamental stages that will be passed through to set up a supplier assessment model for Cordis. It is based on the Critical- To-Quality Tree, as described in paragraph 2.2.3, and serves as the starting point for the definition of the sub-questions.

Figure 7: The conceptual model

16 2.3.3 Sub-questions

The research question will be answered with the help of four sub-questions, which are based on the conceptual model. Moreover, the critical success-factors which have been outlined in figure 5 can be clearly recognized:

•

“Both hard criteria and soft criteria should be measured”: sub-question 1 and 2

•

“Measure and weigh the different criteria”: sub-question 3 and 4

The requirement “compliance with internal and external standards” has been discussed in paragraph 2.2.4, and does not need any further research.

1. Which type of performance measurement method fits best to the situation of Cordis?

To answer this question, different performance measurement methods will be scrutinized. It would go too far to make an inventory of all supplier-rating systems, so we will limit ourselves primarily to the classification used by Van Weele (2002). Other authors that will be studied are, among others, Juran (1951), Jorissen (1994), Thompson (1990), and Saaty (1980).

To find out what is the most appropriate performance measurement method for the situation at hand, first the state of affairs at Cordis should be examined. We need to know for what purpose the method will be used, the type of buying situation it will be used for, and the strategic importance of the products. With the help of this information and the selection- method introduced by Van Weele (1984) it is possible to determine the most appropriate performance measurement method for Cordis.

2. Which criteria should be included in the Supplier Assessment Model for Cordis?

After the most appropriate performance measurement method has been selected, a decision has to be made on the criteria that will be included in the Supplier Assessment Model. In other words, we need to find out what criteria are important for Cordis to assess in order to acquire a good insight in supplier performance. In this context, the know-how of experts within the company is of great importance, since the measurement criteria are to a great extent company specific. A brainstorm session should reveal these company-specific criteria.

Furthermore, intensive use will be made of literature. The assessment criteria applied by different authors will be scrutinized and the importance of these criteria will be discussed. The relevance of literature will, as a rule, decrease as the assessment process becomes more company specific.

Special attention is paid to the impact of supplier assessment criteria on a buying firm’s business performance. For this purpose the study of Kannan and Tan (2002) will be reflected on. They identify the relationship between perceived importance of supplier assesment criteria and business performance. This relationship is taken as a starting point for our assessment model.

3. How can we measure the assessment criteria determined in sub-question 2?

To make the supplier assessment model operational, the criteria, which were determined in sub-question 2, should be quantified. The objective criteria will be measured in an objective manner, i.e. by looking at the number of lots meeting the requirements, divided by the total number of lots during this period. To make the subjective criteria as objective as possible, a checklist will be drawn up so that everyone will assess these criteria in the same manner.

An inevitable subjective element in the supplier assessment model is fixing the maximum points that can be obtained for every criterion, or in other words, the determination of the importance that should be attached to the different criteria. To reduce the subjective element, the ranking of the overall importance of the different criteria, which is obtained in pursuance of the brainstorm session, will be used as supporting information in the decision-process.

4. What weights should be assigned to the different categories?

Another subjective factor in a supplier assessment system is the ascertainment of the

appropriate weight to each category. Some categories are considered more important than

others, and this should be expressed with the help of the weight assigned to them. The

Supplier Quality Assurance department and the Supplier Management department will have

to take the final decision on this topic.

17

2.4 Methodology

In this paragraph the design of the research process will be outlined. In paragraph 2.4.1 the type of research is described, after which the data collection methods and data sources are discussed. Finally, the research planning and the research methods used will be presented.

2.4.1 Type of research

Scientific vs. Practical research

The difference between scientific research and practical research is that the latter recognizes a concrete demonstrable client (or clients) who has a need for knowledge, while scientific managerial research contributes to the general knowledge base of business administration.

The difference is based on the distance between the research and a certain client or group of clients (De Leeuw, 1996: p.74). This research is carried out for two demonstrable clients: the Supplier Quality Assurance department and the Supplier Management department of Cordis.

The objective is not to contribute to the general knowledge base of business administration, but to provide the client with company-specific knowledge on supplier assessment.

The knowledge-products this research project has to produce are:

•

An analysis of the current supplier rating system of Cordis (paragraph 2.2.2);

•

An analysis of the internal and external rules and regulations concerning supplier assessment (appendix 3);

•

Insight in different supplier performance measurement methods (chapter 3);

•

An analysis of the criteria that have to be measured to gain insight in supplier performance (chapter 4);

•

A measurement method for the above-mentioned criteria (chapter 5)

•

A framework for the assessment of service suppliers (chapter 5).

The result of the research will be reported to the departments of Supplier Quality Assurance and Supplier Management.

Problem solving research

De Leeuw (1996: p.77) distinguishes five different types of research, namely:

1. Pure scientific research 2. Society relevant research 3. Policy-relevant research 4. Policy-supporting research 5. Problem solving research

The prototype of scientific practical research is problem-solving research. Problem-solving research is the creation of knowledge about necessary, or in any case sufficient measures to solve management problems. It is, unlike policy-supporting research, integral. This research project will be problem solving of nature. It aims at giving an answer to the complete need for knowledge in order to solve the problem in terms of a working supplier assessment system (De Leeuw, 1996: p.201).

2.4.2 Data collection and data sources

Research data will be collected by means of both desk research and field research. Desk research exclusively makes use of documents. It concerns, among other things, literature study and analysis of written documents. Written documents are for example books,

magazines, reports, protocols and archives. These are all examples of primary data sources.

A report from another research project is an example of a secondary data source. Both primary and secondary data sources will be used during this research project.

Another distinction that can be made with regard to desk research is that between internal

and external data sources. Examples of internal data sources are company documents

18 (procedures, protocols, etc.) and company specific literature. External data sources are not company specific. Both internal and external sources will be applied.

Interviews, brainstorm sessions, and questionnaires are examples of field research, and they will all be used in the course of this research project. Interviews vary to the extent in which they are structured. A distinction should be made between open versus closed interviews (unstructured versus structured interviews respectively). Furthermore, one can distinguish individual and group interviews. For this research primarily individual, relatively open (semi- structured) interviews will be used, where the interviewed persons have the possibility to put forward matters that can bring about insight for the interviewee.

Brainstorming is a technique with which ideas can be generated. The definition of

brainstorming is “a means to acquire in a short time a great number of ideas from a group”

(Rawlinson, 1981). The important aspects of this definition are threefold: a great number of ideas, a group of people, and a short time. Not all ideas that are generated are necessarily useful but all ideas are acceptable during a brainstorm session.

The table below represents for every sub-question the methodology and the data sources that will be used to answer the question.

Sub- question

Methodology Data sources Chapter

1 Desk and field research Literature, internal data sources, semi-structured interviews

3

2 Desk and field research Literature, brainstorming, questionnaire

4

3 Desk and field research Literature, semi-structured interviews 5 4 Desk and field research Semi-structured interviews 5

Table 1: Overview of the methodology and data sources of the research

For the analysis of the current situation with regard to supplier assessment, internal data sources will be applied (Cordis Franchise Method). Moreover, semi-structured interviews with people from the departments directly involved are arranged, i.e. Supplier Quality Assurance, Supplier Management, and Materials Management. The company-specific information required to answer the first sub-question (Which type of performance measurement method fits best to the situation of Cordis?) is acquired through the same interviews. Besides, to answer this question internal data sources and literature will be used. With regard to the different types of supplier performance measurement methods we will use the classification described by Van Weele (2002). Van Weele (1984) will also be consulted to select the most appropriate supplier performance measurement method for Cordis.

To answer the second sub-question, (Which criteria should be included in the supplier assessment model for Cordis?) first a literature review will be conducted, in order to find out what criteria are considered essential in a supplier assessment system. After that, we will focus on the company-specific criteria. For this the group decision-making model for supplier rating (Muralidharan et al., 2002) will be used. A brainstorm session, followed by a

questionnaire in which the participants of the brainstorm session are asked to rank the criteria in order of importance, will have to bring to light the most important Cordis-specific criteria.

The third and fourth sub-question (How can we measure the assessment criteria, and what weights should be assigned to the different categories?) are developed in chapter 5. Both literature and semi-structured interviews will be used to find an answer to these questions.

Literature will be used, where possible, to define the criteria, after which by means of

19 interviews information will be collected on how the criteria can be measured and what weight should be assigned to the different categories.

2.4.3 Research planning

The time schedule of the research, which is represented in table 2, is drawn up on the basis of the different stages that can be distinguished in Process Excellence. Process Excellence is a research method used by Cordis, and practically corresponds to the Diagnosis, Design, and Change (DOV) model of De Leeuw. It is a method to work in a structured way on

improvement, guided by the Six-Sigma approach. This is an approach where one constantly strives to decrease the number of deviations, to achieve continuous and pioneering

improvements. With this approach Cordis tries, among others, to decrease the number of mistakes, waste and rework. In the next paragraph both the DOV-model and the Process Excellence-method will be scrutinized.

Research phase Start date End date

Define June 23rd 2003 August 1st 2003 Measure August 4th 2003 October 17th 2003 Analyse October 20

2003 November 7th 2003 Improve / implement November 10th 2003 December 12

2003 Control December 15th 2003 December 29th 2003

2.4.4 DOV vs. Process Excellence

The DOV-model

The DOV-model is a problem-solving method, and can be divided into three phases, namely:

diagnosis, design and change. These phases do not always follow each other in time, but for convenience’s sake we will discuss them in this order.

The diagnosis process can be explained as a transformation of a signal in a diagnosis, in other words, a transformation of a problem situation in a management problem. With “signal”

we mean a request for help. In this research the signal has been described in the reasons for research (paragraph 2.2.1).

The design-phase encloses developing the diagnosed problem (or the mess of problems) into a concrete solution, and in the change-phase the designed problem has to be implemented.

Here there is question of an operational problem, for here the organization has to change. It is the transformation of a design (a solution) in a concrete system (De Leeuw, 2000: p.297).

Process Excellence

Process Excellence is a method to improve a process, and consists of five phases: Define, Measure, Analyse, Improve / Innovate and Control. In table 3 an overview of the objectives of the different phases is presented, but first every phase will be shortly discussed.

Define

In the define-phase customer issues and requirements are being identified. Then the internal processes need to be identified which “manage” these issues and requirements. The trigger- question one can ask in this phase is: “What are the customer expectations of the process?”

The objective is “to define the project’s purpose and scope and get background on the process and customer.”

Measure

20

In the measure-phase information is gathered to determine how well the process performs in

reality. The trigger-question one can ask is: “What is the frequency of defects?” The objective

is “to focus the improvement effort by gathering information on the current situation.”

21 Analyse

The analyse-phase concerns the research of the process and the identification of the main issues of the problem. The question one can ask here is: “Why, when, and where do defects occur?” The objective is “to identify deep causes and confirm them with data.”

Improve / Innovate

This phase is about generating solutions and developing an implementation plan. “How can we fix the process?” The objective is “to develop, try out, and implement solutions that address deep causes.”

Control

In the control-phase the new process is being standardized and controlled to secure that the desired improvements are permanent. The question one can ask here “How can we make the process stay fixed?” The objective is “to use data to evaluate both the solutions and the plans; to maintain the gains by standardizing the process; and to anticipate next steps.”

Phase Goal

1. Define Define purpose and scope

2. Measure Focus improvement effort by gathering information on the current situation

3. Analyse Identify root causes and confirm them with data

4. Improve / Innovate Try out and implement solutions that address root causes 5. Control Evaluate the solutions and the plan, maintain the gains by

standardizing the process and outline steps for ongoing improvements

A comparison: DOV vs. Process Excellence

The three phases that can be distinguished in the DOV-model can be linked to the phases of the PE-model. Table 4 shows which phases correspond with each other.

De Leeuw (DOV) Process Excellence

Diagnose Define

Measure Analyse

Design Improve

Change Innovate

Control

Table 4: Comparison of the DOV-model and the PE-model

The first three phases of Process Excellence are a more detailed description of the Diagnose-

phase of De Leeuw. The Improve-phase (PE) can be compared to the Design-phase of De

Leeuw, and Innovate, a sub-phase of Improve, can be compared with the last phase of the

DOV-model, i.e. the Change-phase. Process Excellence adds one more phase, namely the

Control-phase. On a high aggregation-level one could state that both research methods

distinguish more or less the same phases.

22 The difference between both methods can be found in their background. The DOV-model is based on management-theories, while Process Excellence has a more or less practical background. Process Excellence is not so much a theory as a collection of techniques that can be used during a certain phase of the research. The DOV-model is less detailed than Process Excellence and can be seen as a research structure within which the researcher determines which theories apply to solve the problem.

On the client’s request Process Excellence-techniques will be used for this research project.

Examples of specific techniques used so far are the SIPOC, the Voice-Of-the-Customer, and

the Critical-To-Quality Tree.