André Luís Rocha da Silva Costarepositorium.sdum.uminho.pt/bitstream/1822/39287/1/André... · 2019. 1. 1. · Universidade do Minho Escola de Engenharia André Luís Rocha da Silva

Universidade do MinhoEscola de Engenharia

André Luís Rocha da Silva Costa

outubro de 2015

Quality Costs Analysis: Case Study in the Automotive Industry

Andr

é Lu

ís R

ocha

da S

ilva

Cos

ta Q

ua

lity

Co

sts

An

aly

sis:

Ca

se S

tud

y in

th

e A

uto

mo

tive

In

du

stry

U

Min

ho|2

015

André Luís Rocha da Silva Costa

outubro de 2015

Quality Costs Analysis: Case Study in the Automotive Industry

Universidade do MinhoEscola de Engenharia

Trabalho efetuado sob orientação do Professor Doutor Paulo Alexandre da Costa Araújo Sampaio

Dissertação de Mestrado Mestrado Integrado em Engenharia e Gestão Industrial

ii

DECLARAÇÃO

Nome: André Luís Rocha da Silva Costa

Endereço eletrónico:[email protected]

Número do Bilhete de Identidade: 13938600

Título da dissertação: Quality Costs Analysis: Case Study in the Automotive Industry

Orientador: Professor Doutor Paulo Alexandre da Costa Araújo Sampaio

Ano de conclusão: 2015

Designação do Mestrado: Mestrado Integrado em Engenharia e Gestão Industrial

DE ACORDO COM A LEGISLAÇÃO EM VIGOR, NÃO É PERMITIDA A REPRODUÇÃO DE QUALQUER PARTE

DESTA DISSERTAÇÃO

Universidade do Minho, ___/___/______

Assinatura:

mailto:[email protected]

iii

ACKNOWLEDGMENTS

This project would not be possible without the contribution of some people, to whom I want to thank for.

Firstly, I would like to thank my supervisor, Professor Paulo Sampaio, for the useful pieces of advice and

for the willingness regarding this project.

Secondly, I would also like to thank Delphi Braga for the opportunity to learn by doing this project in an

industrial environment.

I am most grateful to my Family, who was always there for me. Thank you Mãe, Pedro, João and to you

Pai, I Wish You Were Here.

Finally, I am also thankful to my Friends for their support and assistance whenever I needed and for those

who share with me this experience.

v

ABSTRACT

The main goal of this Project was to analyze with resource to Quality Tools the Current Warranty Process

and the Warranty Claims since 2012 until the First Quarter of 2015, in order to recover from Supplier the

Quality Costs regarding defective materials.

This dissertation was developed in the Automotive Industry in Delphi Automotive Systems Portugal S.A.

Currently, the Quality Costs regarding each Warranty Claim are unknown by Delphi, setting a fixed value

of 1000 € for each Warranty issue, that is not valid at Supplier’s eyes, considering it does not reflect the

real costs of each claim and it is much higher than the price that Delphi paid for each component.

Although, Delphi has a Zero Defects Policy and did not pay for No Quality regarding defective components,

is being charged by Customer. Delphi Braga wants to be the first company in Delphi group to recover

Quality Costs from every Supplier, in order to raise awareness and to ensure that every Supplier is

committed to Delphi Policy and to minimize the current Quality Costs.

According to the Literature, one of the principles of Quality Continuous Improvement is based on the

assumption that every decision, in special those which are taken by Quality Team, should rely on the use

of Quality Tools, for the purpose to discover the critical root causes to be solved and to implement

corrective actions, ensuring the Quality Standards. The higher is the short-term investment in Costs of

Good Quality as Prevention and Appraisal Costs, the less are the long-term costs of Poor Quality

throughout the Supply Chain as Internal and External Failure Costs, ensuring a win-win strategy for all

stakeholders and nurturing business partnership towards Delphi requirements of being known among

Customers as their best Supplier, surpassing their highest expectations.

A rigorous analysis was undertaken to 7696 Warranty Claims from 2012 until the First Quarter of 2015,

concluding that there are 153 complaints confirmed by supplier analysis as defective material

corresponding to a total of 80.724.31€. Given that, the Quality Costs were associated to each Warranty

Claim, creating a cost-oriented Database with all the Customer Debit Costs per device and when added

to the Analysis Costs, the amount that Delphi has to recover from Supplier was calculated.

KEYWORDS

Warranty Claims, Quality Tools, Quality Costs, Cost Recovery, Database

vi

vii

RESUMO

O objetivo principal deste projeto é a análise, com recurso às Ferramenta da Qualidade, do Atual Processo

de Garantia e das Reclamações ao abrigo da Garantia desde 2012 até ao 1º Quadrimestre de 2015,

para recuperar Custos de Qualidade junto do fornecedor relativos a defeitos de material.

Esta dissertação foi desenvolvida na Indústria Automóvel na Delphi Automotive Systems Portugal S.A.

Atualmente, os Custos da Qualidade relativos a cada Reclamação de Garantia não são conhecidos pela

Delphi, tendo-se fixado um valor de 1000 € para cada defeito de Garantia, o que não é válido do ponto

de vista do fornecedor, uma vez que não reflete os custos reais de cada reclamação e é muito superior

ao preço que a Delphi paga por componente fornecido. Embora a Delphi tenha uma Política de Zero

Defeitos e não paga por componentes fornecidos defeituosos, tem de pagar ao Cliente. A Delphi Braga

quer ser a primeira empresa no grupo Delphi a recuperar Custos de Qualidade de todos os Fornecedores,

de forma a consciencializar e a assegurar que cada Fornecedor está comprometido com a Política da

Delphi para minimizar os atuais Custos de Qualidade.

De acordo com a Literatura, um dos princípios da Melhoria Contínua da Qualidade é baseado no

pressuposto de que todas as decisões, em especial aquelas que são tomadas pela Equipa de Qualidade,

devem depender das Ferramentas da Qualidade com o propósito de descobrir as causas-raiz críticas a

serem resolvidas e as medidas corretivas a serem implementadas, garantindo os Padrões de Qualidade.

Quanto maior for o investimento a curto-prazo em Custos de Boa Qualidade, menor são os custos a

longo-prazo da Má Qualidade ao longo da cadeia de abastecimento, assegurando uma estratégia

mutuamente benéfica para todos os intervenientes e alimentando parcerias comerciais, tendo em vista

os requisitos da Delphi, de ser conhecida entre os seus clientes como o seu melhor fornecedor,

superando as suas maiores expectativas.

Foi realizada uma análise a 7696 Reclamações de Garantia desde 2012 até 1ºQuadrimestre de 2015,

concluindo que existem 153 reclamações confirmadas pelo Fornecedor, como sendo defeitos de

material, correspondendo a um total de 80.724.31€. De forma a materializar este valor, os Custos da

Qualidade foram associados a cada Reclamação de Garantia, criando uma base de dados orientada aos

custos, onde são apresentados todos os Custos Debitados pelo Cliente por aparelho e quando somados

aos Custos de Análise, é obtido o valor que a Delphi deve recuperar.

PALAVRAS-CHAVE

Reclamações de Garantia, Ferramentas e Custos da Qualidade, Recuperação de Custos, Base de Dados

viii

ix

CONTENTS

Acknowledgments ............................................................................................................................... iii

Abstract............................................................................................................................................... v

Resumo............................................................................................................................................. vii

List of Figures ..................................................................................................................................... xi

List of Tables .................................................................................................................................... xiii

List of Abbreviations .......................................................................................................................... xv

1. Introduction ................................................................................................................................ 1

1.1 Background......................................................................................................................... 1

1.2 Objectives ........................................................................................................................... 2

1.3 Research Methodology ........................................................................................................ 2

1.4 Structure of Dissertation ...................................................................................................... 5

2. Literature Review ........................................................................................................................ 7

2.1 Quality Concept ................................................................................................................... 7

2.2 Quality Complaints .............................................................................................................. 7

2.3 Quality Costs ....................................................................................................................... 9

2.4 Quality Tools ..................................................................................................................... 12

2.4.1 Pareto Diagram ......................................................................................................... 13

2.4.2 Flow Charts ............................................................................................................... 13

2.4.3 Cause-and-Effect Diagram .......................................................................................... 14

2.4.4 Histogram ................................................................................................................. 15

2.4.5 Check-List ................................................................................................................. 15

2.4.6 Scatter Diagram ........................................................................................................ 16

2.4.7 Control Chart ............................................................................................................. 17

2.4.8 5 Whys ...................................................................................................................... 17

2.4.9 Focus Group .............................................................................................................. 19

3. Case Study ............................................................................................................................... 21

3.1 Company Presentation ...................................................................................................... 21

3.1.1 Delphi Group Introduction .......................................................................................... 21

3.1.2 Identification and Locations in Portugal ...................................................................... 23

3.1.3 Delphi Braga History and Background ........................................................................ 24

x

3.1.4 Products and Customers Portfolio .............................................................................. 25

3.1.5 Quality and Excellence Guidelines .............................................................................. 26

3.1.6 Production Flow ......................................................................................................... 27

3.2 Problem Description and Current Situation ........................................................................ 28

3.2.1 Delphi Quality System ................................................................................................ 28

3.2.2 DPS – Delphi Problem Solving ................................................................................... 30

3.2.3 Defects Categories ..................................................................................................... 31

3.2.4 Quality Costs Recovery Analysis ................................................................................. 32

3.2.5 Warranty Flow............................................................................................................ 34

3.3 Warranty Claims Analysis .................................................................................................. 37

3.3.1 General Analysis ........................................................................................................ 38

3.3.2 Supplier Liability Analysis ........................................................................................... 41

3.3.3 Root Cause Analysis .................................................................................................. 46

3.3.4 Quality Indicators ....................................................................................................... 49

3.4 Quality Costs ..................................................................................................................... 51

3.4.1 Analysis and Handling Costs ...................................................................................... 51

3.4.2 Database with Customer Debit Costs per Device ........................................................ 52

3.4.3 Warranty Claims Costs ............................................................................................... 54

3.4.4 Supplier Perspective .................................................................................................. 56

4. Conclusions .............................................................................................................................. 59

4.1 Final Considerations .......................................................................................................... 59

4.2 Limitations ........................................................................................................................ 63

4.3 Future Work ...................................................................................................................... 63

Bibliography ..................................................................................................................................... 65

Appendix I – Cost Recovery Procedure .............................................................................................. 69

Appendix II– Radio Components ....................................................................................................... 71

Appendix III - Warranty Follow-up Instructions .................................................................................... 73

Appendix IV – Root Cause Categories ................................................................................................ 75

Appendix V – Cost Recovery Formal Message .................................................................................... 77

xi

LIST OF FIGURES

Figure 1 – Action Research Model adapted from O´Brien (1998) ........................................................ 3

Figure 2 - The Quality Costs adapted from Krishnan et al. (2000). ..................................................... 10

Figure 3 - Pareto Diagram ................................................................................................................. 13

Figure 4 - Flow Chart ........................................................................................................................ 13

Figure 5 - Cause-and-Effect Diagram ................................................................................................. 14

Figure 6 - Histogram ......................................................................................................................... 15

Figure 7 - Check-List ......................................................................................................................... 16

Figure 8 - Scatter Diagram ................................................................................................................ 16

Figure 9 - Control Charts ................................................................................................................... 17

Figure 10 - 5 Whys Analysis .............................................................................................................. 18

Figure 11 - Focus Group Representation (Source: www.groupquality.com) ......................................... 19

Figure 12 - Delphi Worldwide. ........................................................................................................... 21

Figure 13 - Major Business Segments ............................................................................................... 22

Figure 14 – Organization Values ....................................................................................................... 22

Figure 15 – Delphi locations in Portugal ............................................................................................ 23

Figure 16 – Delphi Industrial Complex in Braga (Source: www.googlemaps.com) ............................... 24

Figure 17 – Main Product Families, Production Volume versus Revenue ............................................ 25

Figure 18 - Main Customers, according to Revenue Percentage ......................................................... 26

Figure 19 - Plastics and Infotainment Production Phases ................................................................... 27

Figure 20 - The Quality Checkpoints .................................................................................................. 28

Figure 21 - DPS Flowchart ................................................................................................................ 30

Figure 22 - The 3 Defects Categories within Supply Chain ................................................................. 31

Figure 23 - Belt Reject Cost Recovery ............................................................................................... 33

Figure 24 - Customer Impact Cost Recovery ...................................................................................... 33

Figure 25 - Warranty Cost Recovery .................................................................................................. 33

Figure 26 - Current Warranty Flow Procedure .................................................................................... 35

Figure 27 - Warranty Analysis Flow .................................................................................................... 36

Figure 28 - Warranty Claims per Year ................................................................................................ 38

Figure 29 - Warranty Claims Status Percentage ................................................................................. 39

Figure 30 - The Customers with Most Complaints .............................................................................. 39

xii

Figure 31 - Warranty Claims Liability ................................................................................................. 40

Figure 32 - Pre-Analysis Margin of Error ............................................................................................ 40

Figure 33 - Warranty Claims Liability in Percentage ........................................................................... 40

Figure 34 - Representation of Law Sets Theory .................................................................................. 41

Figure 35 - The Complaints that are Supplier responsibility per year .................................................. 42

Figure 36 - Pie Chart with the Complaints Status ............................................................................... 42

Figure 37 - Pie Chart with the Approved Complaints .......................................................................... 43

Figure 38 - Pareto Diagram representing the Defective Components .................................................. 44

Figure 39 - Complaints Root Causes ................................................................................................. 45

Figure 40 - The PCB Defective Hole .................................................................................................. 46

Figure 41 - Micro-Section Defective Hole Analysis .............................................................................. 46

Figure 42 - Root Cause Analysis - 5 Whys for Occurrence .................................................................. 47

Figure 43 - Root Cause Analysis – 5 Whys for No Detection ............................................................... 47

Figure 44 - Laser Diode Output in Field Pattern COD Sample and Normal Sample ............................. 48

Figure 45 - Root Cause Analysis with Fishbone Diagram .................................................................... 48

Figure 46 - Scatter Chart with Warranty Period versus Odometer ....................................................... 50

Figure 47 – Complaint Analysis Duration .......................................................................................... 50

Figure 48 - Warranty Claim Duration ................................................................................................. 51

Figure 49 – Database Extract with Customer Debit Costs per Device ................................................. 53

Figure 50 - The Debit Notes Distribution ............................................................................................ 55

Figure 51 - Supplier Perspective ........................................................................................................ 57

Figure 52 - Cost Recovery Procedure ................................................................................................ 69

Figure 53 - Radio Components .......................................................................................................... 71

Figure 54 - Warranty Follow-up Instructions ....................................................................................... 73

xiii

LIST OF TABLES

Table 1 - Service Partner Analysis Debit Note .................................................................................... 52

Table 2 - Delphi Analysis and Handling Costs .................................................................................... 52

Table 3 - Defective Components and Radios Part Numbers................................................................ 52

Table 4 - Warranty Costs by Defective Component ............................................................................. 54

Table 5 - Delphi Debit Notes per Root Causes ................................................................................... 56

Table 6 - Root Cause Categories ....................................................................................................... 75

xv

LIST OF ABBREVIATIONS

AOI – Automated Optical Inspection

BOM – Bill of Materials

CAR – Corrective Actions Record

CEO – Chief Executive Officer

COPQ – Costs of Poor Quality

COQ – Costs of Quality

CSE – Customer Satisfaction Engineering

DPRTS – Delphi Problem Resolution and Tracking System

EOS – Electrical Overstress

ESD – Electrostatic Discharge

ISO - International Standards Organization

NOK – Not Ok

NTF – No Trouble Found

P/N – Part Number

PLC – Public Limited Company

PQM – Product Quality Manager

RMA – Return Material Authorization

ROW – Rest of the World

SAP - Name of company providing Materials & Finance & Focus Factory

SMT - Surface Mount Technology

SQE – Supplier Quality Engineering

THT – Through Hole Technology

Quality Costs Analysis: Case Study in the Automotive industry

1

1. INTRODUCTION

This chapter presents the Background that motivated this project, the Objectives that led the way, the

Research Methodology that best suited the addressed objectives for this project and the Structure of

Dissertation.

1.1 Background

Nowadays, satisfying the customers’ needs is not enough to any company who wants to thrive or even

survive in the market. The competition is ferocious due to the environment of uncertainty and crisis, as

Charles Darwin as cited in Wellburn (1996) once said “It is not the strongest or the most intelligent who

will survive but those who can best manage change”. Therefore and quoting Feigenbaum as cited in

Reed, Lemak, & Mero (2000) “Quality is what customer says it is”. That is why the focus of every company

must be, the Satisfaction of customers’ needs, surpassing their highest expectations through an efficient

use of the available resources. In this context, the Quality as a measure of Excellence is more important

than ever in every industry.

According to Talay, Calantone, & Voorhees (2014) the Automotive Industry is characterized by an

unstoppable change of competitive dynamics and consumer preferences. There is an increasing

competitive pressure from the incumbents and newcomers, therefore the automakers continually need

to improve their ability to manage their supply chains and product development processes. One way of

keep up with the competition is giving more responsibility to the suppliers, so “Across many industries,

companies especially in automotive sector increasingly give more responsibility to their suppliers to design

and produce innovative, high quality products at a lower and competitive cost” (Ayaǧ & Samanlioglu,

2014). Especially in this industry, there is the need to be ahead of the competition through innovation

and customer’s needs fulfillment, therefore the Service Quality is mandatory, such as the continuous

management of Quality Costs.

Quality Costs are inherent to every process in the continuous pursue for Quality. The aim of every

Company is to provide the best Quality practices with the lowest Costs, however is not easy as might

think. Therefore, the complexity of the problems requires the use of Quality Tools in accordance with

Hagemeyer, Gershenson, & Johnson (2006) to aid the organizations in the analysis of information, to

identify the critical factors that should be improved. Some authors referred to Quality Tools as a leverage


2

in the competition, that is strategical or even a mean to grant survival, being an essential asset in every

organization (Kathuria & Davis, 1999).

According to Sousa, Aspinwall, Sampaio, & Rodrigues (2005), the success of process improvement is

due to Quality Tools, which are an important advantage to identify and solve problems. Although there

are a great variety of Quality Tools, the most popular are the seven basic Quality Tools. During this project,

the Quality Tools are fundamental to identify and to minimize the Quality Costs within a company that

supply infotainment for the Automotive Industry. The focus is on customer complaints, since is one of the

best ways to evaluate Customer Satisfaction (Bolfing, 1989). However, a special attention is given to

Warranty Claims caused by supplier components, since in the actual process there is not a share of the

risk related to every complaint. Delphi is covering all the Warranty Costs, even when the problem root

cause is provoked by defective supplied components. Therefore, it was done an analysis of every warranty

claim since 2012 until the first quarter of 2015 helped by Quality Tools that allow consolidating the Quality

Costs of warranty that should be recovered by Delphi.

1.2 Objectives

The main goal of this project is to analyze the current Warranty Process and the Warranty Claims since

2012 until the first quarter of 2015, in order to recover the Quality Costs regarding supplied defective

material. For that purpose, the partial goals proposed are the following:

a. Quantify all the warranty costs from the moment that the device is returned from the final customer

to Delphi’s Braga Plant for root cause analysis until suppliers problem confirmation;

b. Create a database where is possible to obtain the cost per defective device;

c. Increase Supplier acceptance rate of Warranty Claims;

d. Increase the Quality Standards in every step of the process.

1.3 Research Methodology

The selection of the topic for this project was based in an analysis of forces, needs and personal interests.

The company where this project was developed had a special interest in this topic, due to the potentialities

of this study in their daily life. This project was structured with the mentoring sessions of the college tutor

and with the support of the company where was developed. On the other hand, the ongoing review of the

literature as dissertations, international articles and essays, played a major role to achieve the proposed

goals.


3

As referred before, the main goal of this project is to quantify the quality costs regarding every warranty

claim which root cause was provoked by supplier, in order to recover those costs and to simplify this

process for future complaints.

The methodology of research used during the project is the Case Study and the Action Research, since

they complement each other, as it is shown below.

According to Tereso (2011) the Case Study develops a detailed and intensive knowledge, which is very

useful to answer to “WH-questions”, as why, what and how. The methods which are used by this

methodology of research are interviews, surveys, content analysis and direct observation. This method is

a valid way of exploring and questioning the existing theory, by formulating new hypothesis.

According to O´Brien (1998) the Action Research is knowing by “learning by doing”, since a group of

people, the researcher and the employees, identify the problem, try to solve it within the planned time,

see if they were successful and in case of dissatisfaction, they iterate the process all over again. The

researcher interacts with the employees and acts in the work field, where is able to participate in real

world situations and solve real problems, since people learn better when they are able to apply their

knowledge in practical situations. According to Gerald Susman as cited in O´Brien (1998) this

methodology follows 5 phases within each research iteration. The 5 phases, as presented in Figure 1,

are: Diagnosing Phase, Action Planning Phase, Taking Action Phase, Evaluating Phase and the last one,

is the Specifying Learning Phase, where is done a sum up of the main targets accomplished as well, the

identification of the problems that were not solved.

Figure 1 – Action Research Model adapted from O´Brien (1998)

Diagnosing

Action Planning

Taking Action

Evaluating

Specifying Learning


4

Through this methodology, the main goal of this master’s thesis is the search, analysis and

implementation of a well-documented system to quantify all the costs that exist in every warranty claim

in order to make the supplier take responsibility and to assume the costs recovery. Simultaneously, it is

conducted an intensive academic research in scientific articles from prominent journals of the field, books

and master’s thesis related with the main topic, the Quality Costs. After that, it is done a critical review

of the literature, which main goal is the sum up of relevant information about the topic, in a way that is

possible to obtain all the knowledge that is important and crucial for the project.

On the first phase of the methodology, it is done a diagnostic of the current situation of the warranty claim

process, which identifies the problems that contributes to the quality costs and as well, the information

that should be considered in the contact with the supplier. For that, the Quality Tools have an important

role in the reduction of Quality Costs or Costs of no Quality. The Pareto Diagram is used in the analysis

of the company’s documentation regarding the number of warranty claims per year, per customer, per

supplier and per product according to the defect that took place. Then, the Ishikawa Diagram and the 5

Whys plays an important role to discover the causes of the two most frequent detected defects aided by

Focus Group, which allows a debate constituted by employees from different departments of the plant to

obtain reliable information concerning the origin of the detected defects. Through direct observation of

the costs related to each warranty claim are taken in consideration and quantified, such as costs of

transportation, analysis and tests (impact, temperature, programming) of the root cause of the problem

that in the majority of the cases involves the disassemble of the device to locate the defective part, time

per worker and per diagnostic machinery wasted in each claim.

On the second phase of the methodology, it is identified all the possible solutions to improve the current

process of the warranty claims, according to the problems that were found before. In this stage, the

project is focused on the path that brings more successful outcomes by doing an action planning to

implement on the next phase.

On the third phase, there is the Action Planning Implementation and as consequence, the outcomes that

came from it. There is the need to write down every result and evaluate its performance.

On the fourth phase, it is done an analysis and the discussion of the results obtained and also a

comparison between the current situation and the proposed one, in a way that are shown the advantages

that results from the suggested changes. The Quality Costs are a good indicator of the project success,

since the main goal of the project is to create a well-documented system where it is presented all the

Quality Costs related to warranty claims which the suppliers are responsible for. Therefore, the Quality

Costs decrease through supplier cost recovery.


5

At least, on the fifth phase, it is done the collection of all obtained results from the proposals that were

successfully implemented and the presentation of future work suggestions.

1.4 Structure of Dissertation

The present dissertation is organized in 4 chapters.

In this chapter, it is presented the Background that motivated this project, the Objectives that led the way,

the Research Methodology that best suited the addressed objectives for this project and the Structure of

Dissertation.

In the second chapter is done the review of the literature about the relevant topics for this project,

beginning with the Quality Concept, complementing different points of view. Afterwards, it is made an

explanation about the importance of Quality Complaints, of the Quality Costs due to Good and Bad Quality,

followed by the Quality Tools and their role in identifying and solving Quality problems.

The third chapter begins by the Company Presentation, followed by the Problem Description and Current

Situation, the Warranty Claims Analysis and the concerned Quality Costs.

The final chapter presents the Final Considerations regarding the project development, the limitations

that appeared along the way and the future work that can be done to improve the current results to

increase the quality in the company daily routine.


6


7

2. LITERATURE REVIEW

In this chapter a review of the literature is presented about the relevant topics for this project.

Firstly, an approach to the Quality concept is made, complementing different points of view.

Afterwards, it is made an explanation about the Quality Costs and its importance.

Finally, there are presented the Quality Tools: Flowcharts, Cause-Effect Diagrams, Pareto Analysis,

Histograms, Check Sheets, Control Charts, Scatter Diagram, 5 Whys and Focus Group with the purpose

to identify the root causes of warranty claims and the associated Quality Costs.

2.1 Quality Concept

The concept of Quality is wide and changes from situation to situation and depends on the perception of

each individual about what is the quality of a product or a service.

According to Oakland (1995) the term quality includes customers’ requirements, their needs and

expectations regarding the product or service. The author refers that the concept has been used for many

Quality Gurus in many ways. Juran (1988) defines Quality as “fitness for use”, Crosby (1979) says that

Quality is “conformance to requirements”, Gilmore (1970) defines “Quality is the degree to which a

specific product conforms to a design or specification” as while Deming sees it in way that should fulfill

the customer present and future needs. On the other hand, to Delphi, the company where this project

was developed, Quality consists on being known among customers as their best supplier, surpassing

their highest expectations.

2.2 Quality Complaints

In order to fulfill customer requirements, the organizations need to assure that every product or service

is provided under a high level of quality that is consistent and aims customer satisfaction, regardless the

size and the company sector of activity (Tzelepis, Tsekouras, Skuras, & Dimara, 2006). One of the main

goals of every company is to offer a competitive product having in mind the relation between quality and

the costs associated (Cabral, Colaço, & Guerreiro, 2002). The company interest to satisfy customer needs

and expectations regarding quality, trust, safety and market last trends are aspects very important to

customer (Tsim, Yeung, & Edgar, 2002). It is considered customer satisfaction when the product satisfies

or exceeds the expectations. Satisfaction is also defined as the effect caused by product added value on

customer when the features surpasses customer expectations (Wahid & Corner, 2009).


8

According to Gerson (1998) the relation between sales, the service, satisfaction and the profit margin is

direct, given that if the customer is satisfied, he is going to buy again and the company sells more,

increasing the sales volume. Many organizations believe that their products and services satisfy fully all

the customers. Although, these cases are not so often, due to the market demanding criteria and

uncertainty, getting more difficult to satisfy every customer regarding the price, quality, time and

performance.

The organizations depend on customers, therefore it is more important than ever to know their present

and future need to be able to satisfy them. Quality must be always a guarantee and ongoing process, in

order for the company to thrive and to evolve. The improvements should be done at the level of the

product, process and customer satisfaction and does not mean necessarily that there are problems going

on (Piskar & Dolinsek, 2006).

Nowadays, the quality improvement programs have been adopted by the companies to develop their

several processes. The company’s strategic decision making regarding the quality continuous

improvement should be supported on customer’s feedback (Pyon, Woo, & Park, 2011). Customer

feedback is a source of innovation for the process improvements and optimization. Consumer complaints

have an important role in every company, through the insight of customers’ feedback, the organizations

get to know the strengths and the improvement opportunities of their products (Zeithaml & Bitner, 2000).

Complaints Management System is important to set the quality goals (McAlister & Erffmeyer, 2003). The

authors consider that the problems issues concerning the purchased products and the product returns

are the main factors that contributes to customers complaints and there are only a few opportunities

given to customers to express their satisfaction, which are essential to keep business relationships. The

best way to keep a business relationship with a customer is to serve him well from the very beginning.

According to Sampaio, Saraiva, & Rodrigues (2011), the organizations have a mixture of motivations to

implement the complaint management system, in order to increase and promote marketing and

companies competitiveness.

According to Battaglia (2014) the probability to keep the commercial relationships is higher on the

customers that complaint than on the customers that do not show their satisfaction. Every company

should manage willingly each complaint, considering that when customer complains, he has high

expectations towards the quality of the purchased product or service, in contrast with the customers that

do not spend time with complaints.

According to Bosch and Enríquez (2005), the complaints are inevitable due to the different perceptions

of each customer, what is acceptable to one is not acceptable to another, that is why the customer sets


9

the level of quality. The quality indicators are important to perceive the customer satisfaction and to follow

up the complaints and the respective solutions, nurturing the relation with customer and showing them

that the organizations care about their opinion and that they are important to their development.

The defects represent high costs for the organizations and may jeopardize the companies’ image,

resulting in rework and customer dissatisfaction costs. Although, the effort that are made regarding

customer satisfaction does not always translate into good results due to customer expectations, that is

why knowing the customer behaviors are so important (Chen & Cheng, 2010). According to Arter (2003),

non-conformity occurs when the requirements are not attended. As soon as any defect is detected, there

is the immediate need to implement corrective actions to solve it and to prevent the reoccurrence.

The process of complaints management should identify the root causes of the non-conformities, of the

corrective actions and to prevent their reoccurrence, reducing the waste and the ineffectiveness costs

through continuous improvement. That is why, the quality team must be focused on problems resolution

and quality improvement (Chen & Cheng, 2010).

The authors stands that quality and efficiency increasing has a direct impact on the products selling prices

and on the organization competitiveness. The focus should always be on customer satisfaction,

decreasing the time on complaints management and sales returns, as well on the cases of rework due

to lack of quality, warranty process and products replacement (Anderson, Fornell, & Rust, 1997).

Muffatto and Panizzolo (1995) suggested the implementation of a customer satisfaction plan, based on

many key factors as the customer focus, the product and process design, continuous improvement cycles,

organizational changes, top management commitment and monitoring. Although, the preliminary results

showed that customer satisfaction plan implementation is also a problem for the organizations committed

to quality continuous improvement, that is why it is important the commitment to the plan and to the

organizational needs, through multiple feedback cycles (Muffatto & Panizzolo, 1995).

2.3 Quality Costs

There are many authors who have introduced a Quality Cost Models, which can be classified into five

groups of generic models (Schiffauerova & Thomson, 2006). The P-A-F (Prevention-Appraisal-Failure),

Crosby’s Model related to conformance and non-conformance, Opportunity or Intangible Cost Model,

Process Cost Model and ABC (Activities Based Costing) about value-added and non-value-added.


10

Since, most Costs of Quality models are based on P-A-F classification, the focus will be on the P-A-F Model

that stands for Prevention, Appraisal and Failure Costs. The PAF scheme has been almost universally

accepted for quality costing, which failure costs can be further classified into two subcategories: internal

failure and external failure costs.

According to the American Society for Quality Control (ASQC), as cited in Krishnan, Agus, & Husain

(2000), states that the quality costs measure the costs associated with the achievement or non-

achievement of the product or service quality to satisfy the requirements established by the company and

its contracts with customers and society. Horngren as cited in Krishnan et al.(2000) states that “The

costs of quality are those costs that are incurred to prevent a shortfall in quality and a failure to meet

customer requirements, as well as costs incurred when quality does in fact fail to meet customer

requirements”. On the other hand, Harrington’s study as cited in Krishnan et al.(2000), pointed that the

Costs of Good Quality presented in the Figure 2, have the goal to reduce the costs linked to Poor Quality.

Figure 2 - The Quality Costs adapted from Krishnan et al. (2000).


11

As shown in the Figure 2, the Quality Costs can be divided in 4 categories: Prevention Costs, Appraisal

Costs, Internal failure Costs and External Failure Costs. The Prevention Costs are associated to quality

planning, evaluation and improvement. These costs are planned and incurred before actual operation.

The Appraisal Costs are related to supplier’s and customer’s assessment of purchased materials,

processes, intermediates, products and services to assure conformance with the specified requisites. The

Internal Failure Costs takes place when the results of work do not reach designed Quality Standards and

are detected before transfer to customer takes place. Finally, the External Failure Costs happen when

products or services are not able to reach design Quality Standards nevertheless they are not detected

until after the transference to customer (Tsai, 1998).

The basic suppositions of the P-A-F model are that investment in prevention and appraisal activities will

reduce failure costs, and that further investment in prevention activities will reduce appraisal costs (Tsai,

1998). Therefore, the best way to prevent cost of incoming material inspections is to choose the suppliers

that provide the high-quality material or to help suppliers to improve their quality system by establishing

Quality Control Policies or assurance systems to avoid the excessive in-process inspections due to the

complex design of the manufactured products. Through a shared work by all the stakeholders is possible

to identify all the cost saving opportunities.

As mentioned above, the Quality Costs are important to ensure the Excellence of the product, nevertheless

they are also very important to maintain the Customer Satisfaction according to their specifications.

According to Gerson (1998), the organizations should not forget the cost for the customer that includes

the time and the money that he spends with the product purchasing, that might not corresponds to his

expectations.

There are several problems that have caused cost-of-quality approaches to fail. In general, the Quality

cost analysis looks at the company’s costs, not to the customer’s costs. However, the manufacturers and

the suppliers are definitely not the only stakeholders who incur Quality-related costs. The customers suffer

quality-related costs too. If a manufacturer sells a bad product, the customer faces significant expenses

in dealing with that bad product. Every company is interested in the success of its economy. It is

imperative, therefore, that the concept of quality cost thinking and leadership is extended to every step of

the Supply Chain, integrated into business strategic thinking and into daily organizational activities (Aniza,

Wang, & Rieger, 2013) .

The Customer Satisfaction is related to the product and service organizations, focused on an output

perspective on how customers evaluate performance. The key internal Quality practices of product versus

service organizations (team management, process orientation, and customer orientation) influence


12

Customer Satisfaction and as well business results. Therefore, for product organizations, internal quality

practices influence Customer Satisfaction and business results through an organization customer

orientation. For service organizations, customer and process orientation impact customers directly and

team management has a direct impact on business results, which organizations with a Quality foundation

are in a better position to adopt a customer orientation (Manal, Joo, & Shouming, 2013).

2.4 Quality Tools

One of the principles of quality continuous improvement is based on the assumption that every decision,

in special those which are taken by the quality team and by top management, should rely on the use of

Quality Tools (Paliska, Pavletic, & Sokovic, 2007). These tools are useful in control, analysis and data

organization, that are relevant to the decision making process.

During the years, many authors as Tarı́ & Sabater (2004) and Juran & De Feo (2010) have identified a

great variety of Quality Tools and other techniques to improve the companies programs of Quality

Improvement.

According to Tari & Sabater (2004), seven basic quality tools were identified: Flowcharts, Cause-Effect

Diagrams, Pareto Analysis, Histograms, Check Sheets, Control Charts and Scatter Diagram. In addition

to the seven basic quality tools, the Focus Group and 5 Whys are also addressed.


13

2.4.1 Pareto Diagram

The Pareto Diagram is a distribution of attributes grouped by categories (Figure 3), which objective is to

identify and visualize the categories most meaningful in a certain analysis in a fast and easy way. The

Pareto Diagram is one of the seven most used Quality Tool (Montgomery, 2009).

Figure 3 - Pareto Diagram

2.4.2 Flow Charts

The Flow Chart is a scheme that presents all the phase of a process or procedure (Figure 4). The Flow

Chart should identify the flow of the process, as well the interactions between the process phases. One

of the great advantages of the organizations that use this tool for representing their process, is the fact

that it can help identifying the potentials control points and process improvements (Montgomery, 2009).

Figure 4 - Flow Chart


14

2.4.3 Cause-and-Effect Diagram

Every time that a company identifies a defect, an error, or a problem, the potential causes of that defect

and the consequences resulted from it should be analyzed. When the causes are not clear, the Cause-

and-Effect Diagram utilization allows an easier comprehension in the relation of the potential causes of

the produced defect (Figure 5). This tool should be used by a team that want to improve the quality of

the process with the purpose to identify the main problematic areas.

There are 7 steps to build a Cause-and-Effect Diagram and to implement corrective actions, which are:

1. Define the problem or consequence that is analyzed;

2. Create a team to analyze the problem or the consequence (Improvement Quality Team), that

should be a multidisciplinary team and should analyze the problem using Brainstorming when

analyzing problems in manufacturing processes ;

3. Build the diagram central line;

4. Specify the main problems by causes categories and link them to the central line;

5. Identify the main causes and classify them in categories. More causes can be added later;

6. Sort the identified causes from the ones that seems to have more impact on the problem or

defect;

7. Implement Corrective Actions.

In the analysis of the problem or defect, the cause should be classified accordingly to the

machines, the material, the measures and the collaborators (Montgomery, 2009).

Figure 5 - Cause-and-Effect Diagram


15

2.4.4 Histogram

The Histogram is identical to Bar Chart, used many times to represent the distribution of frequencies

(Figure 6). The distribution of frequencies represents the frequency of each different value occurs in a set

of different values. The bars, in this type of chart, present the data grouped, with the purpose to show

the relation between the variables. The Histogram, is one of the quality tools that allows to verify the

numerical data distribution, especially when the user wants to determine if the analyzed data follows a

Normal Distribution. This tool is very useful to check, in a fast and easy way, the data distribution

(Montgomery, 2009).

Figure 6 - Histogram

2.4.5 Check-List

In order to improve a process it is necessary to collect former data, operational data and current data

related with the process under analysis. The Check-List is very useful to help the process improvement,

in a way that allows to collect all the required data for the analysis and improvement of the process

(Figure 7). Every time that a Check-List is needed it is important to include information as what kind of

data will be collected for the analysis, the date, the analyst name and the process identification

(Montgomery, 2009).


16

Figure 7 - Check-List

2.4.6 Scatter Diagram

The Scatter Diagram is used to identify the potential relations between two variables. If there are data

correlated, then the dots will overlap under the shape of a line or a curve (Figure 8). This quality tool

should be used to pair data; when a dependent variable can have many values for each value of

independent variable; when wants to check if the variables are related as possible causes of problems or

have effects in common (Montgomery, 2009).

Figure 8 - Scatter Diagram


17

2.4.7 Control Chart

The Control Chart is used to study how a process in constant change varies over time (Figure 9). This

chart presents a central line that represents the process average, a top line that represent top limit of

control and inferior line that represent inferior limit of control. These lines are determined from recorded

data. Through the comparison between the current data to the chart lines it is possible to draw

conclusions, like if the data distribution is under control or if it is stable (Montgomery, 2009).

Figure 9 - Control Charts

2.4.8 5 Whys

According to Islam (2006), the 5 Whys is an iterative interrogative technique used to explore the cause-

and-effect relationships underlying a particular problem. By repeatedly asking the question “Why”, the

layers of symptoms can be peeled away which can lead to the root cause of the problem. The primary

goal of the technique is to determine the root cause of a defect or problem by repeating the question

"Why?" Each question forms the basis of the next question. The "5" in the name derives from an empirical

observation on the number of iterations typically required to solve the problem. This technique is a

common problem solving process, used to analyze and solve quality problems as showed in Figure 10.

There are four major parts of the process of using this technique to solve a problem:

Grasp the Situation

Cause Investigation

Problem Correction

Prevention through Error proofing


18

Grasp the Situation

During the first part of the process there are three steps to grasp the situation:

Identify the Problem;

Clarify the Problem;

Locate the Point of Cause (PoC).

Cause Investigation

In the second part of the process there are three steps to identify the root cause:

for the specific problem;

to discover why the problem was not detected;

to discover why the problem occurred.

Problem Correction

In the third part of the process it is explained how to correct the problem:

Take specific action to correct the problem. At a minimum, short-term temporary

measures are required to protect the customer.

Prevention Through Error proofing

In the fourth part of the process it is explained how to prevent the problem reoccurrence:

Take specific actions to make sure the problem can not reoccur, typically through error

proofing;

Capture Lessons Learned.

Figure 10 - 5 Whys Analysis


19

2.4.9 Focus Group

The Focus Group is used to gather a collective point of view from several participants at the same time

(Figure 11). It can be used to test if a certain hypothesis is true or to get feedback on some topics defined

by the investigator (Morgan, 1997).

This Quality Tool has some adversities, as can be time consuming, the moderator can induce biases, and

the participants’ personalities can dominate the outcomes. On the other hand, there are several benefits

related to this technique, as the needs definition and clarification, providing deep insights, allowing the

needs prioritization. Furthermore, it is easy to use, allows a high level and immediate feedback rate and

facilitates to test the concepts under discussion.

Figure 11 - Focus Group Representation (Source: www.groupquality.com)


20


21

3. CASE STUDY

In this chapter, it is presented the Company where this project was developed, followed by the Problem

Description and Current Situation, the Warranty Claims Analysis and the concerned Quality Costs.

3.1 Company Presentation

In this chapter, the company where this master thesis was developed it is presented. Initially, in order to

raise awareness about the company roots and way of working, an approach to Delphi Group is needed,

moving on to Delphi Automotive Systems – Portugal S.A. In this context, the focus was in Braga Plant

going from its History and Background, Products and Customers portfolio, Quality and Excellence

Guidelines, until the Production Flow.

3.1.1 Delphi Group Introduction

The company where the project took place was in Delphi Automotive Systems - Portugal SA in Braga, one

of the world biggest suppliers of automotive parts, from Delphi Group headquartered in Troy, Michigan in

United States of America. The group is presented in 270 localizations by 32 countries and counts with

more than 170 thousand employees with revenues of 12 thousand millions euros per year.

Delphi was created by General Motors and became fully independent in 1999. Since then, is known as a

diversified and worldwide organization, with a strong corporate culture enhanced through innovation,

dedication, and commitment to the customer, a vital key to Delphi reputation as one of the world's premier

automotive suppliers, focused to be recognized by the customers as their best supplier.

As the largest and most diversified supplier of automotive parts, Delphi can provide every vehicle

manufacturer customers with global, single-point sourcing capability and systems tailored to meet their

specific needs as is possible to see in Figure 12 the presence of Delphi across the globe.

Figure 12 - Delphi Worldwide.


22

The Delphi Group is organized in 5 major business segments (Figure 13) which are part of the Technology

Portfolio, focused on solutions to customers’ problems, which result is market-driven and strategic

product lines. According to O’Neal, Delphi CEO, (2014), “Delphi is uniquely focused on creating

shareholder value by delivering Safe, Green and Connected products that meet the needs of our

diversified customers through flawless execution and operational excellence” (Figure 14).

Figure 13 - Major Business Segments

Figure 14 – Organization Values

- Zero Fatalities;

- Zero Injuries;

- Zero Accidents.

Visionary products that protect global natural resources.

New products that keep people

connected regardless of time and location.


23

3.1.2 Identification and Locations in Portugal

In Portugal, Delphi has three plants located in Braga, Castelo Branco and Seixal. Furthermore, has a

Technology Center in Lumiar as the Figure 15 shows.

Figure 15 – Delphi locations in Portugal

Since this project was developed in Delphi Automotive Systems – Portugal in Braga, the focus was on

this plant which is responsible for 700 employees and generates revenues of 285 million of euros per

year. The product portfolio is constituted by auto-radios, navigation systems and antennas devices as part

of Electronics & Safety division which provides components, systems and software for both passive and

active safety, security, comfort and infotainment, as well as other vehicular electronic controls. Delphi’s

philosophy is to be known among its customers as the best supplier which can surpass the customer’s

highest expectations.


24

The company is formed by four buildings with a site floor space of 32.921 m2 (Figure 16).

The Building 1 is where the final products manufacturing take place. The Building 2 corresponds to the

production of plastic parts that are assembled later on the radios and antennas. The shipping warehouse

is also located in the Building 2, meanwhile the Building 3 and 4 is where the raw material storage is

done.

Figure 16 – Delphi Industrial Complex in Braga (Source: www.googlemaps.com)

3.1.3 Delphi Braga History and Background

Initially, Delphi Braga was known from 1965 as Grundig, becoming part of Delphi Group in 2003, when

was acquired. The company foundations goes back to 1965, when Max Grundig, founder of the largest

European radio manufacturer known as Grundig AG, bought for nine hundred “contos” the land where

the plant would be working. Back to those days, the first radio to be assembled was a “Transonette”

followed by the success of the “Heinzelmann”, a radio receptor that would be the responsible for a sales

boom for years to come, until 1996 when Grundig had the worst year in terms of financial results.

http://www.googlemaps.com/


25

In 2003, when the multinational American group Delphi bought Grundig Car Intermedia System, the plant

located in Braga became nominated as Delphi Grundig until 2011, when the company became Delphi

Automotive Systems - Portugal S.A. until today.

In 2009, the products portfolio was expanded from Car Radios to Antennas. Meanwhile, in 2010 began

the production of plastic components that supply Braga Plant and other Delphi plants across the globe.

3.1.4 Products and Customers Portfolio

Delphi has a wide products portfolio with high manufacturing complexity in order to obtain products with

brand new features and cutting edge technology to reach and to surpass the customer requirements and

expectations. The development of these products is done by the Manufacturing Technical Centers in

America, Europe and China. However, the majority of projects allocated to Braga are developed in Poland

and Germany.

The range of products manufactured in Braga is constituted by Auto Radios, Navigation Systems,

Reception Systems and Plastic Components. In terms of production volume the Reception Systems and

the Auto Radios represents 65% and 30%, respectively. However, what concerns to revenue, the Auto

Radios are the most valuable group to the company, representing 80% of the incomes, while the

Receptions Systems worth only 12% (Figure 17).

Figure 17 – Main Product Families, Production Volume versus Revenue

Production Volume Revenue

65%

12%

30%

80%


26

When it comes to its customers, Delphi works with the main vehicles manufacturers around the planet

as Volkswagen Group, General Motors, FIAT Chrysler Automobiles, Ford Motor Company, BMW Group

and some other recognized and well-known brands. The Volkswagen Group is the Delphi Braga Top Client,

being responsible for 69% of the annual revenue, where is included VW, Audi, Skoda and Porsche,

followed by FIAT Chrysler Automobiles and Volvo Group, both with 7% as can be seen in Figure 18.

Figure 18 - Main Customers, according to Revenue Percentage

3.1.5 Quality and Excellence Guidelines

Delphi is strongly committed to be known among its customers as their best supplier, surpassing

customer’s highest expectations. This purpose is only achievable, having Excellence as keystone and as

requirement to Quality. Only through continuous improvement, led by the organizational values and best

practices is possible to thrive in an environment as competitive and ferocious as the Automotive Industry.

That being said, as a company that produce more than 1,4 millions of Auto Radios and 3,2 millions of

Antennas in an annually basis, Delphi Braga is certified by ISO 9001:1994 Quality Management System

Certification, by ISO/TS 16949 Quality Management System Certification for Automotive Industry and by

ISO 14001 Environmental Management System Certification to meet customer requirements efficiently

and effectively in a sustainable way.


27

3.1.6 Production Flow

As already mentioned, the production flow is divided between the Building 1 and 2. The Building 1 is

responsible for the Infotainment referred as Final Products and the Building 2 manufactures the plastic

components to be assembled on the Final Products. The plastic components production flow begins with

the Injection Molding phase, then the pieces goes to Painting Process and then come together in Final

Assembly. The Infotainment production flow begins with Surface Mount Technology (SMT) in which

electronic components are inserted automatically onto the surface of Printed Circuit Boards (PCB), then

Through Hole Technology (THT) and Sticklead phase is initiated when components are placed manually

on PCBs. At last, the Final Assembly brings together the Trimplate or Cover (Auto Radio and Antenna

frontal panel) and other parts as the Cd Player Mechanism with the main board. After that, the Final

Product is ready for Testing, Packing and Shipping (Figure 19).

Figure 19 - Plastics and Infotainment Production Phases

1. Injection Molding

2. Painting Process

3. Final Assembly

1. SMT

2. THT

3. Final Assembly


28

3.2 Problem Description and Current Situation

The problem that motivated this project was the quality costs that Delphi has with warranty claims caused

by supplier. Until now, there are only a couple of suppliers who agreed to share the warranty costs which

they are responsible for. They agreed to share the warranty costs since they caused epidemic situations

in the past, by providing defective components to Delphi that damaged several devices on the road. On

the other hand, the majority of suppliers that are responsible for less complaints individually were never

charged before for warranty costs, due to contract obligations or due to the fixed cost that Delphi set for

every warranty claim of 1000 € that is much higher than the price that Delphi paid for each component.

Although, Delphi has a Zero Defects Policy and states that did not pay for defective components which

are being charged by Customer. Delphi Braga wants to be the first company in Delphi group to recover

quality costs from every supplier, in order to aware every supplier to Delphi Policy, to minimize the current

quality costs and to avoid their reoccurrence in the future.

3.2.1 Delphi Quality System

Quality plays a major role in Delphi. Throughout the production flow, there are several checkpoints to

ensure the quality of the entire process as shown in the Figure 20. In Delphi everyone is strongly

committed to zero defects policy, in order to guaranty customer satisfaction.

Figure 20 - The Quality Checkpoints


29

The Quality process begins at supplier facilities, before the shipment to Delphi, supplier needs to ensure

that every component that were provided to Delphi, follows the strict quality standards demanded by the

automotive industry. When the material arrives to Delphi, a quality control is done to detect any non-

conformity through visual inspection. If the material is under the specifications, the order is approved and

follows to the supermarket, where the material is stored per Part Number (P/N) and follows to the

production lines. In case of detection of any non-conformity at the order arrival, the material is blocked

and a complaint is issued to supplier with a sample of the defective material. The “skip-lot” statue is

changed for that part number until corrective actions are taken by supplier to remove the problem and to

ensure the quality of the future orders.

The implementation of corrective actions is monitored closely by Delphi and if there are costs, they are

debited to supplier. When the corrective actions are not undertaken within the expected time frame, the

blocked material is inspected and in some cases, the pieces can go under rework, depending on the

production needs regarding that component. If there are enough pieces of the contained part number in

the supermarket to satisfy the production orders, the blocked material is returned to supplier or destroyed

by Delphi.

There are 3 quality checkpoints inside the production area, located after each production phase. The

first checkpoint is the Optical Automated Inspection, where the electronic connections and the surface

joint points of the circuit boards assembled at SMT are checked. Afterwards, there is the Manual

Production stage, followed by the second quality point, which checks the conformity of the assembled

components. At last, the final quality stage, is also known as the last line of defense before the products

are shipped to customer, being this quality checkpoint decisive to comply with customer requirements

and to avoid 0 km and warranty claims. It is important to mention, that the components damaged during

the production flow are disposed, while those components detected as supplier liability are collected and

returned.


30

3.2.2 DPS – Delphi Problem Solving

Regarding Quality issues, DPS is the method used by Delphi to solve problems and to make decisions

when the problem root causes are unknown as presented in Figure 21.

Figure 21 - DPS Flowchart

The steps taken by DPS are:

1. Ensure the problem description to be analyzed in detail;

2. Contain the problem, whenever it reveals necessary;

3. Analyze and eliminate the root causes through corrective actions implementation;

4. Control and Standardize to Prevent Problem Recurrence.


31

3.2.3 Defects Categories

At Delphi, the defects are divided in 3 major categories, they are Belt Reject, Customer Impact or 0 km

and Warranty (Figure 22). Following the supply chain, the first category includes all the defective material

detected in Delphi Production Area and it is rejected by the Conveyor Belt. The Customer Impact includes

the defects detected at customer manufacturing plant. Finally, the Warranty Issues includes all the defects

that happens at final customer hands on the road.

Figure 22 - The 3 Defects Categories within Supply Chain

In what concerns to defect liability, it is always a complex process to find out the source of the problem

whatever the defect category. However, having in consideration that the source of the problem is always

present on the supply chain, the further the defect is detected within the chain, the harder it gets to prove

it is someone fault. Therefore, despite the accuracy of analysis techniques used to find out the defect root

cause, business relation plays an important and decisive role, regarding trust, transparency and good

faith issues. Having that in mind, it is easier to identify the responsible of the problem in a belt reject

defect, than in a warranty issue, since almost all of the components that are part of the radio at final

customer hands travelled a long way within the supply chain. The higher the complexity to find out the

root cause responsible, the more probable it gets that Delphi must cover all the costs and damages

caused to customer, underestimating the risk sharing policy inherent to any defect, that is why Delphi

must be committed to quality excellence and to avoid the occurrence of any kind of defect.


32

3.2.4 Quality Costs Recovery Analysis

When along the supply chain is detected any problem regarding material defects issues, Delphi Problem

Solving is put in place. When there are costs involved, Cost Recovery Process is initiated (Appendix I).

Cost Recovery is defined as the process of communicating with a supplier, documenting costs and

expenses such as inspection, sorting, re-work, lost production units, replacement units and downtime

expenses as they are incurred by Delphi as the result of a supplier nonconformance, and collecting such

costs and expenses from supplier.

As referred before, there are 3 types of costs that Delphi can be impacted in case of supplier delivers

non-conform parts:

1. Belt Reject - impacts Delphi plant and happens when the problem escapes from Supplier detection

and it is found at Delphi manufacturing plant. If no significant impact is caused a value of 40€ per line

pull is applied. In case of significant impact, then a specific calculation it is done.

2. Customer Impact or 0 km - impacts Delphi Customer and occurs when the defect escapes from

Supplier detection, pass Delphi manufacturing plant and is detected at Delphi Customer manufacturing

plant. If no significant impact is caused a value of 700€ per case it is applied In case of significant impact,

then a specific calculation it is done.

3. Warranty or Field: impacts Final Customer and happens when the defects escapes from supplier

detection, passes Delphi manufacturing plant, is not found at Delphi Customer manufacturing plant and

is detected by the Final Customer. If no significant impact is caused a value of 1.000€ per case it is

applied. In case of significant impact, then a specific calculation it is done.

In order to perceive the suppliers acceptance for each type of cost, it is essential to analyze the Costs

Recoveries made by Delphi since 2012. In the Figure 23, Figure 24 and Figure 25, there are comparisons

between the amount of money proposed by Delphi and the amount agreed by Suppliers in each cost

recovery for each type of defect.


33

Figure 23 - Belt Reject Cost Recovery

Figure 24 - Customer Impact Cost Recovery

Figure 25 - Warranty Cost Recovery

36.404 €

12.500 €

- €

5.000 €

10.000 €

15.000 €

20.000 €

25.000 €

30.000 €

35.000 €

40.000 €

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33

Belt Reject Cost Recovery 2012-2015 1ºQ

COST RECOVERY VALUE (€) SUPPLIER AGREED (€)

34.975 €

27.058 €

- €

10.000 €

20.000 €

30.000 €

40.000 €

1 2 3 4 5 6 7 8 9 10 11

Customer Impact Cost Recovery 2012-2015 1ºQ


173.483 €

249.530 €

69.393 €24.090 €

- €

50.000 €

100.000 €

150.000 €

200.000 €

250.000 €

300.000 €

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Warranty Cost Recovery 2012-2015 1ºQ



34

According to the Figure 23, Figure 24 and Figure 25, since 2012 there were 33 cases in Belt Reject, 11

cases in Customer Impact and 15 cases in Warranty. In Belt Reject the Costs Recoveries are done in a

monthly basis and the majority of the activities charged to supplier are inspections, sorting, re-work and

rarely there are costs of downtime expenses, since Delphi has a safety stock of 7 days and only on

exceptional circumstances the defect on the supplied component, mainly electronic parts, can not be re-

worked. Although, Delphi Suppliers are from all over the world, the supplier lead time is not superior to

the safety stock duration. Through the observation of Figure 23, it is possible to notice that the line from

the Cost Recovery Value and the line from Supplier Agreed are most of time overlapped, excepting on the

3 highest values, where the supplier pays only a fraction of what Delphi asks for.

When comes to Customer Impact Cost Recoveries, the lines are also overlapped, excepting on the highest

value, that Supplier pays less of what Delphi asks for as presented in Figure 24. The Warranty Cost

Recoveries values of Figure 25 are much higher than the other defects, since these ones are done only

for the epidemic cases that affects many units, setting aside the suppliers with less complaints

individually. The Cost Recoveries done so far were only for the Suppliers that Delphi established an

agreement with. Although, even under agreement, the differences between the proposed costs to recover

and what the supplier agreed are considerable, mainly on the highest amounts. The differences are high,

since the agreements sets a fixed percentage that supplier must accept of the total amount that Delphi

proposes. Since Delphi has an agreement only with two suppliers, a deeper analysis of all the Warranty

Claims is essential to determine the real Quality Costs linked to each Warranty Claim, in order to minimize

those costs.

3.2.5 Warranty Flow

The focus of this project is to quantify the costs associated to each warranty claim caused by supplier

component and to recover those costs. Therefore it is important to know the Current Warranty Flow

Procedure, from moment that the device fails at Final Customer until the Customer Debit Note (Figure

26). The warranty period begins when the vehicle has a license number and is sold to a Final Customer,

who has 2 years, from that moment, to claim any problem regarding the vehicle. In case of any problem

related to a radio supplied by Delphi, the warranty process begins when the customer is contacted and

opens a Delphi Problem Resolution Tracking System (DPRTS). This database is the communication

channel between Delphi and its Customers, regarding complaints. When the case is opened, the

Customer has the choice to define the priority, which can be formal or informal. The formal cases have

an impact on Delphi Quality Metrics, having visibility for all the group, that is why Delphi has an immediate


35

response to allocate resources to these kind of issues. When the customer realizes that the case it is

solved within the expected time, the complaint priority changes to informal until the process is closed.

When the device fails at Final Customer vehicle, the device it is collected by the dealer that sold the car.

Then, the unit is sent to Delphi Service Partners to analyze the problem root cause, through visual

inspection analysis and electrical tests. The analysis done by Service Partners is also designated as Pre-

Analysis, since the analysis is superficial compared to the analysis done in Delphi Laboratory. According

to the analysis outcome, there are two options. If the problem root cause is already known and there is

data from previous analysis to support the results obtained, the device does not need to go to Delphi

Laboratory for further analysis. On the other hand, if the analysis is inconclusive and the root cause has

not appeared before in other units, the device is sent to Braga and further analysis are done. In both

situations, the analysis aim is to discover the defect liability that can lie on Supplier, if it is a material

defect, on Delphi process or on Customer. When the defect, is customer responsibility, it means that the

device was damaged during customer process or was damaged by final customer. In case of material

defect, the device should be sent to supplier attention for further analysis to confirm the diagnostic.

Although, no matter what are the Supplier conclusions, as soon as customer liability scenario is discarded,

Delphi must pay all the warranty costs regarding the complaint.

Figure 26 - Current Warranty Flow Procedure


36

As referred before, the Warranty Analysis has place in Delphi Laboratory when the Service Partner analysis

is inconclusive. Currently, only 5% of warranty claims are being analyzed in Delphi Laboratory, due to

logistics and handling issues, Delphi relies on outsourcing Service Partners that collect and analyze the

defective radios majority. Although, the warranty claims resulting from future projects, will be analyzed in

Braga, which will demand an increasing on the available capacity.

In contrast with Service Partners, Delphi analysis requires a deeper approach, beginning with the device

inspection on Production Lines, followed by a work bench tests to verify the device functionalities as

presented in Figure 27. When those tests are not effective, the device it is under furthers analysis of

temperature and vibration to confirm the failure mode. Afterwards, the device is disassembled for visual

inspection and electrical measurements are done leading to ABA Cross-Check Test, allowing the problem

root cause identification or leading to further analysis (Appendix II – Radio Components). The analysis

conclusions result in an 8D Report, Eight Disciplines of Problem Solving, that is sent to Supplier, if the

material defect is confirmed, for further analysis to confirm Delphi diagnostic. The Supplier will also

elaborate an 8D Report, presenting the analysis done to the component and the results.

Figure 27 - Warranty Analysis Flow


37

The Warranty Flow is a complex process, having in consideration also the Warranty Follow-up Instructions

present in the Appendix III. The steps that need to be followed, from the moment that the defective

component is analyzed in Delphi until the Supplier 8D elaboration is time consuming. The process to

complaint the defective component to Supplier demands the access to many different databases with

dispersed information as DPRTS, Corrective Actions Record (CAR), SAP and Warranty_Supplier_File (SQE

internal document with the warranty claims analyzed in Braga). If there was a database that centered all

the needed piece of information, the information flow would be improved significantly, since it is important

to have an integrated image of the 3 type of problems. The same root cause can affect the same

component P/N or the same device P/N across the 3 types of problems, that is why it is important to

see the big picture and to discover similarities between problems to have a concerted action in their

solution.

3.3 Warranty Claims Analysis

Since, the main target of the project is to quantify and analyze all the cost related to every warranty claim,

in order to recover those costs, which root causes were caused by defective supplied components. Having

always in mind Delphi zero defects policy and customers’ requirements, the supplier must take the

responsibility for its actions and share the risk with Delphi. The analysis of data from 2012 until the 2015

first quarter, allows to establish a model for Delphi future warranty claims, increasing Delphi cost recovery

process efficiency and effectiveness. This process it is completely different from what Delphi having been

doing till now, since quantifies the costs of every and each warranty claim, recovering the money debited

by each customer in every warranty claim not only through the suppliers that maintain agreements with,

nevertheless also with the other suppliers that are responsible for warranty claims that never took on their

share of liability.

There was also the need to categorize the root cause analysis description and the problem description

fields of each warranty claim into short standard categories that are understandable and easy to identify

and transform into data used for graphical analysis.


38

3.3.1 General Analysis

As referred before, the focus is on every Warranty Claim from 2012 until the first quarter of 2015, since

this timeframe, from a cost-benefit perspective is the most likely to enhance suppliers receptivity to cost

recovery through the analysis of certified reports available on Delphi database DPRTS, besides, that period

is well representative of the company record and behavior when comes to warranty claims and what

should be the approach in the years to come.

The data was extracted selectively from DPRTS, the system used by customer to open and follow

complaints that can be “customer impact” or “warranty”, which must be solved through corrective actions

taken by Delphi along with its suppliers, when the situation demands it. According to the data collected,

there are 7696 warranty claims since 2012, namely 6822 cases that are closed and 874 open cases

that are pending for corrective actions, as shown in Figure 28 the number of closed cases are decreasing

from 2012 until first quarter of 2015, although is still unclear if that will be the tendency of the rest of the

year. When comes to the open cases, the distribution is not so clear, being the first quarter of 2015, with

263 open cases, the period with more ongoing cases, in contrast with 2012, that presents 171 cases.

Figure 28 - Warranty Claims per Year

2561

2170

1706

385

171245 195 263

0

500

1000

1500

2000

2500

3000

2012 2013 2014 2015 Q1

Warranty Claims per Year

Close

Open


39

What concerns to the cases percentage distribution, the pie chart is the best way to represent it into

slices, as presented in the Figure 29 the ongoing cases are a minority with 11% of the distribution against

the cases already closed that sums 89% of the distribution.

Figure 29 - Warranty Claims Status Percentage

When the warranty claims are distributed according to customer focus, the Pareto Diagram plays an

important role, displaying the customers, as letters from A to F due to confidential issues, with the most

number of complaints as A, B and C that represents 83% of the total number of complaints, followed by

D, E, F and other customers with the least number of complaints in the Figure 30. The majority of

complaints are concentrated in 3 customers out of 13.

Figure 30 - The Customers with Most Complaints

89%

11%

Warranty Claims Status

Close

Open

39%

62%

83%

93%97% 99% 100%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

0

500

1000

1500

2000

2500

3000

3500

A B C D E F Others

The Customers with Most Complaints

Count Cumulative %


40

Every defect is a serious problem, in terms of quality, resources and money. No matter if it is a belt, a 0

Km or a warranty issue, the obstacle needs to be overcame by Delphi in the first instance, however when

the problem affects the customer and its customers, the problem is even more serious and complex to

solve, therefore the need to find out the problem source is critical to avoid its reoccurrence in the future.

Having that in mind and focus on the closed complaints, the liability of each warranty claim is represented

with a comparison between the Pre-Analysis results and the Final Analysis. According to the Figure 31,

Customer is responsible for the majority of claims, followed by Supplier and by Delphi, respectively. Since,

there are some differences between the Pre-Analysis and the Final Analysis regarding the liability of the

problem, there was the need to test the Pre-analysis margin of error that is 1% as shown in Figure 32,

meaning that pre-analysis match the final analysis almost always.

Figure 31 - Warranty Claims Liability Figure 32 - Pre-Analysis Margin of Error

The pie analysis shows that more than half of the warranty claims with 57% is caused by the customer

or its customers, followed by suppliers with 34% and finally, Delphi with 9% of the cases as shown in

Figure 33.

.

Figure 33 - Warranty Claims Liability in Percentage

3920

2297

605

3945

2266

611

0 1000 2000 3000 4000

CUSTOMER

SUPPLIER

DELPHI

Warranty Claim Liability

Pre-Analysis Final Analysis

99%

1%

Pre-Analysis Margin of Error

Realibility Error

57%

9%

34%

Warranty Claims Liability

Customer Delphi Supplier


41

As referred before, the analysis focus is on the 2297 warranty cases which root cause was caused by

Supplier, and for that, Supplier ID and Component Part Number were needed in order to identify the

supplier and the defective component linked to each root cause. Due to the database lack of information,

those two fields were created and through DPRTS ID, 8D Report Analysis and SAP it was possible to

collect some data as the supplier ID for 1961 cases and the component part number for 1415 cases.

According to the Figure 34, the best way to represent this situation is through Law Sets Theory, illustrating

that Supplier ID is a subset of Supplier Liability and Component Part Number is a subset of Supplier ID,

three essential fields to proceed with the analysis and the costs recovery regarding the 1415 warranty

claims.

Figure 34 - Representation of Law Sets Theory

3.3.2 Supplier Liability Analysis

The 1415 warranty claims are considered as supplier responsibility by Delphi Laboratory Analysis after

disassemble the affected auto radios and extract the defective component that is the root cause of the

observed failures. However, the supplier before accepting the complaint liability wants to receive its

defective component to do a complete analysis, confirming the problem root cause.

Through observation of the Figure 35, there is the distribution of the warranty claims since 2012 until the

first quarter of 2015 divided in those which were accepted or rejected by Supplier Analysis and the ones

without enough data to confirm whether they were accepted or rejected by Supplier. The

complaints considered without enough data are mainly from 2012, the oldest ones, which may indict that

the information were not saved properly or the service partner did not send the defective parts to supplier

analysis, not being able to confirm the root causes of the failures. Only in 2014, strict instructions were

given to the service partners to send the defective components back to the supplier attention to go under

analysis to confirm the root cause conclusion by Delphi Analysis and to save the supplier 8D Reports.

Supplier Liability

2297 Cases

Supplier ID

1961 Cases

Component P/N

1415 Cases


42

That fact, might be why 2015 has the least “no data” complaints. In the case of the accepted complaints,

it is possible to conclude that they were accepted by supplier since 2012, that is why the cost recovery

practice should have been in place earlier.

Figure 35 - The Complaints that are Supplier responsibility per year

When comes to percentage distribution, the Figure 36 provides that information in a pie chart, where 66%

of the complaints were rejected by supplier analysis, refusing the complaints liability. That percentage is

so high, possibly due to the fact that Delphi Laboratory analysis is not as robust as should be. There are

few analysts to analyze all the complaints that arrive to Delphi, although they represent only 5% of all the

complaints regarding Delphi Braga devices. Since there are a lack of resources, the analysts are pushed

by Customer Satisfaction Engineering to analyze the cases as fast as they can, to deliver them to Supplier

Quality Engineering, that are responsible for the communication with the supplier and for the case follow

up. Thus, when the analysis from Delphi is not well founded, the Supplier have more leverage to refuse

the defect liability, since the supplier manufactured the defective component and knows everything about

it. This situation happens also with analysis from the Service Partners, that are superficial and does not

stand a chance when confronted with the analysis undertaken by Supplier, a fact that Delphi can not

control, since the Service Partners follow up all the process.

Figure 36 - Pie Chart with the Complaints Status

0 50 100 150 200 250 300 350 400 450 500

2012

2013

2014

2015

Complaints Status per Year

Rejected Accepted No Data

66%

20%

14%

Complaints Status

Rejected

Approved

No Data


43

The complaints that were recognized by supplier as their responsibility, summed 20% of the cases,

followed by the cancelled complaints due to lack of data with 14% of the cases. This lack of data was

constant during the database analysis, since in the description of the complaint it was indicated that the

8D Report was in attachment, however in reality, it was not. That is why the database should be updated

and more organized, to allow an easy collection of data, to be less time consuming and more Cost

Recovery Oriented.

Proceeding to the accepted complaints, as shown in the Figure 37, there are different status regarding

these complaints, the 93 cases corresponds to the couple of suppliers with whom Delphi has an

Agreement in Place. On the other hand, there are 34 cases from different suppliers, that are less than 5

cases per supplier, a criteria adopted to increase the suppliers receptivity to cost recoveries, however

those costs are postponed for more upcoming cases. Finally, there are 153 cases under the scope which

are electable candidates for costs recovery.

Figure 37 - Pie Chart with the Approved Complaints

As shown in the Figure 38, the pareto diagram allows to observe that the Circuit Board and the Cd Player

are the most defective components affected by complaints, representing 81% of the total cases, followed

by the Bluetooth Moduls, Transistors and ICs. The deffective components were supplied by 6 suppliers,

being responsible by each type of component, exepting the circuit boards, the component with most

complaints, provided by 2 suppliers, one with 63 complaints and the other one with 29 complaints.

Through the diagram observation, all the defective components presented are electronic components,

which is related to ESD, Electrostatic Discharge, sensibility factor, since a mechanical or a plastic part is

more robust. On the other hand, for example a cosmetic defect or a dimension defect on a trimplate, the

auto radio plastic front part, which is a probable defect to happen and easy to detect is completely

93

34

153

Approved Status Situations

Agreement in Place

Less than 5 Cases

Under the Scope


44

different from a electrical component. A electrical part can be supplied already damaged, however the

failure mode only appears under certain conditions of stress such as impact, heat, humidity and dust that

wear out the component and are only felt when the radio is already on the road. Unfortunately, this kind

of situations are hard to detect before the radio arrival to the customer hands, that is why the company

should share this risks and costs with the supplier that supplied the defective component and improve

the process continuosly to prevent this kind of defects.

Figure 38 - Pareto Diagram representing the Defective Components

Although, the damaged electrical components often cause the failure mode appearance after wear out

under certain conditions of stress, there are mechanical or plastic parts mainly small, which could be bad

assembled in the radio interior. That scenario makes the defect difficult to be detected, however there

are inspection techniques as shaking the radio in certain positions to check if there are strange noises

coming from its inside due to loose parts, that were bad assembled or dropped by, that are effective.

However, it can happen also that the CD player gear rack spring or the eject lever spring are bad

assembled or the main slider is deformed, that are only detected in warranty claims, as presented in the

Figure 39. In Figure 39 is showed the 24 root causes linked to the 153 approved complaints in warranty.

Through the chart observation, Via Hole Open Issue and Catastrophic Optical Damage (COD) are the top

root causes in 60 and 22 of the cases under scope, representing more than half of the total cases since

92

32

1410

5

60%

81%90%

97% 100%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

0

10

20

30

40

50

60

70

80

90

100

Circuit Board CD Player Bluetooth Modul Transistor IC

The Defective Components

Count Cumulative %


45

these problems affected several auto radios assembled with supplied components from different models

and production batches, until corrective actions were put in place.

Figure 39 - Complaints Root Causes

In order to transform the root cause description presented in each warranty claim into direct and easy

understandable data, it was defined for this project the need to create Root Cause Categories that will be

used by Delphi from now on, to describe the root causes in future warranty claims (Appendix IV). As

example, the Root Cause description presented in DPRTS for Via Hole Open Issue, the most frequent

cause of the defects observed, is the following: “The customer complaint was confirmed, after 30

seconds after tuning into a station- static noise can be heard that gets louder overtime. The failure is

Intermittent. On Quality Laboratory bench, the symptom was identified, after dump Eeprom, it has verified

that the radio make some noises and also in CD on Front Left speaker. Electrical analyses verified that

the problem is on the circuit of amplifier, one via hole from the main board should measuring zero ohm

60

22

13

9

7

6

5

5

4

4

3

2

2

1

1

1

1

1

1

1

1

1

1

1

0 10 20 30 40 50 60 70

Via Hole Open Issue

Catastrophic Optical Damage (COD)

Containers Contamination

Package Corrosive Elements

Electrical Spikes

Crack at the Bottom of Laser Via

Solder Mask Surface Scratched

Wrong Pffs Autodetection

ASIC P5+ IC Defective

Random FAB Particle

Defective PCB

Deformed Roller

Internal Structure Impaired

BT Chip Defective

C116 Defective

Defective OPU

Disturbed Vcc Bond Wires

Eject Lever Spring Bad Assembled

Gear Rack Spring Bad Assembled

Main Slider Deformed

U1702 IC Defective

U6 IC Progamm

Wafer Fabrication Defect

Wirebonded Leadframe Contact

Complaints Root Causes


46

and is intermittent normally with the value of 16 Ohm. X-ray: The suspected area has been carefully

checked with a microscope, and not only this via hole but a lot hole are defect. Similar cases are known

like CAR 4798. Supplier conclusion last report: Per above micro-section images, we can see there was

some drilling burrs in the hole orifice, which led to the copper plating chemical could not go through the

hole completely, so, somewhere on the hole wall just plating a thin copper”. Having in mind this

description, through Root Cause Categories the information flow is facilitate and the analysis process is

reinforced and more robust, minimizing the Costs of Poor Quality and reinforcing the Good Quality.

3.3.3 Root Cause Analysis

In order to understand in detail the root causes that caused more than half of the warranty claims under

analysis, the focus is on the origin of these root causes to provide a deeper understanding, through some

Quality Tools as the Cause and Effect Diagram by Ishikawa and 5Whys method aided by Focus Group.

The Focus Group was composed by Supplier Quality Engineers, Laboratory Analysts, Supplier, Product

Manager Engineers and Industrial Engineers, since the more diverse the group, the more rich and fruitful

is the feedback gathered to discover the reasons behind each root cause in analysis.

The Via Hole Open Issue is detected by measuring the resistance of the defect location on the PCB

(Printed Circuit Board) showed in the Figure 40 and discover that the AC, Alternating Current, does not

flow in the trace connected by the hole. According to micro-section images in the Figure 41, there are

some drilling burrs inside the hole, since there was some foreign material that made the drill to break,

blocking the copper plating chemical, that could not go through the hole completely. Therefore, the AC

was not flowing since the plating copper on the hole wall was too thin.

Figure 40 - The PCB Defective Hole Figure 41 - Micro-Section Defective Hole Analysis


47

In order to understand and to clarify the root cause analysis occurrence and the reason why this problem

was not detected in time, 5Whys analyses are presented in the Figure 42 and in the Figure 43,

respectively.

Figure 42 - Root Cause Analysis - 5 Whys for Occurrence

Figure 43 - Root Cause Analysis – 5 Whys for No Detection

The corrective actions resulted from the root causes analysis to solve this problem, taken by the supplier,

are the following:

1. Before drilling, the boards go through the rinsing line to clean the surface from any residue;

2. An alarm sounds every time there is a drill broken and the present board is marked;

3. The Work Instructions were improved, as well the Operator skills for this kind of defect.

The second root cause that caused more warranty claims was the Catastrophic Optical Damage (COD),

which is detected by measuring high peaks of AC in the Laser Diode. This happens when the component

has received an electrostatic discharge or a surge current. The consequence is a reduced output light

power as shown in the Destructive Test at Figure 44. In this case, the Laser control circuit tries to keep

the light power and increases the operation current to compensate the lack of power. This increased


48

current in consequence speed up the process of destruction by further deterioration in power reduction

until the Laser Diode is completely dead in Electrical Overstress (EOS).

Figure 44 - Laser Diode Output in Field Pattern COD Sample and Normal Sample

In order, to clarify the root cause analysis and all the possible causes that led to Laser Diode EOS, a

Cause Effect Diagram is presented in the Figure 45.

Figure 45 - Root Cause Analysis with Fishbone Diagram


49

In order to protect from potential ESD, Electrostatic Discharge, pre-damaged Laser Diodes being delivered

to Delphi, the Laser Diode current is measured at the end of the production process. The measured value

is automatically compared with the initial value of the OPU, Optical Pick-Up Unit, and the module is

rejected on deviation.

The corrective actions resulted from the root causes analysis to solve this problem, taken by the supplier,

are the following:

1. Continuous ESD prevention with internal audits to check the individual protective equipment and

the use of ESD protective bags to accommodate the supplied material;

2. Limit the max. Laser Diode current by increasing the line resistor from 4,7 to 5,6 Ohm, which

prevents COD to happen.

Through the use of Quality Tools, as presented above, to discover the critical root causes to be solved, to

implement supplier corrective actions were vital, ensuring the Quality Standards in the beginning of the

Supply Chain. The higher is the investment in Costs of Good Quality as Prevention and Appraisal Costs,

less are the costs of Poor Quality throughout the Supply Chain as Internal and External Failure Costs,

ensuring a win-win strategy for all stakeholders and nurturing the business partnership.

3.3.4 Quality Indicators

The Warranty Period Analysis and the Car Mileage are important quality key indicators, since it allows to

check if there is a tendency in the complaints data distribution of the 1415 cases as supplier liability

according to Delphi analysis, displayed in the Figure 46. The warranty period considered is since the car

was sold until the car breakdown. By observing the scatter chart, it is possible to notice that the majority

of the complaints occurred within the car warranty period of 730 days, what means that the warranty

period time was enough to protect customers, excepting in the 45 cases that were out of warranty,

according to the data. Regarding the car mileage record with an average of 16669 km, when compared

to the highest value of the distribution with 226982 km, the value is relatively low. Through the chart

observation, when the focus is on the cases out of warranty and on the cases with a mileage above the

average, a total of 22 cases, is noticed that there is not a relationship between the car mileage and the

warranty period.


50

Figure 46 - Scatter Chart with Warranty Period versus Odometer

The time that takes to conclude a warranty claim and a complaint analysis is a quality indicator that is

evaluated by the customer and matters to customer satisfaction. Therefore, a deep analysis to the

warranty data regarding the complaint analysis duration and the total complaint duration is imperative.

The Figure 47 shows the complaints sorted from the newest to the oldest, revealing the distribution of

the analysis duration in the complaints, where is an average comparison regarding the analysis done in

Delphi Braga and the analysis done by Delphi Service Partners. According to the chart, in Braga the

analysis duration takes in average 6 days, which is half of the 12 days that the services partners take to

analyze each complaint. The results are surprising, since Delphi Braga has a lack of laboratory analysts,

having in consideration that 3 people analyses all the defective devices send by the customer, whether

are warranty or 0 km issues.

Figure 47 – Complaint Analysis Duration

612

0

100

200

300

0 200 400 600 800 1000 1200 1400

Complaint Analysis Duration

Analysis Duration Average Braga Average Service Partners

226982

730

16669

0

500

1000

1500

2000

2500

0 50000 100000 150000 200000 250000

Day

s si

nce

Car

Sel

ling

until

Bre

akdo

wn

Cars Odometer in Km

Warranty Period versus Odometer

Warranty Complaints Warranty Period Limit Odometer Average


51

On the other hand, the Figure 48 reveals the complaints duration average, which is from the moment

when the customer made the complaint until its completion. By observing the chart, it is noticed that

Delphi Braga takes 21 days in average to process a complaint, in contrast with Service Partners, who

spends 37 days in average to close a warranty claim. Since, all the complaints from new projects will be

analyzed in Braga, these numbers are a good indication of what should be expected, although there will

be a service increasing, what might be need to reinforce the workforce to maintain the customer

satisfaction.

Figure 48 - Warranty Claim Duration

3.4 Quality Costs

In order, to quantify all the costs associated to a warranty claim, there was the need to establish several

contacts within the organizational departments inside Delphi as Customer Satisfaction, Industrial

Engineering, Finances and Supplier Quality Engineering, where this project was being held, to understand

all the steps of the warranty process.

Until now, only in exceptional circumstances as epidemic situations where many radios were affected by

a common root cause, Delphi established an agreement with a couple of suppliers to allow Costs

Recovery. The standard value defined by Delphi for warranty cost recovery per unit is 1000 euros, a fixed

value that is not valid in suppliers’ eyes, since it is not a variable that resulted from real costs calculation.

That is why it is important to calculate every cost involved in each Warranty Claim.

3.4.1 Analysis and Handling Costs

The costs associated to every warranty claim are composed by the Analysis and Handling Costs and by

the Customer Debit Costs per Device. The costs related to the complaint analysis includes the Pre-Analysis

Costs debited to Delphi by Service Partners and the Analysis and Handling Costs done by Delphi Braga.

2137

0

100

200

300

400

500

0 200 400 600 800 1000 1200 1400

Warranty Claim Duration

Warranty Complaint Duration Average Braga Average Service Partners


52

If the complaint is managed entirely by the service partners, the costs associated are separated in 15

euros for the Logistic Costs (radio transportation from customer facilities to service partner facilities) and

in 34 euros for the Pre-Analysis Cost done by service partner, with a total of 49 euros as presented in the

Table 1. In case of the pre-analysis concludes that the root cause of the problem was never seen before,

the radio is send to Delphi Braga for further analysis. The costs associated to Delphi Braga analysis

consists in 15 euros for the Logistic Costs (radio transportation from service partner facilities to Braga),

in 33 euros for the analysis costs (cost for 1h of laboratory analyst time) and in 60 euros for handling

costs (cost for 1h of work done by two Supplier Quality Engineers), that sums up in 108 euros as

presented in the Table 2 plus the 49 euros from the Service Partner Debit Note, with a total of 157 euros.

Table 1 - Service Partner Analysis Debit Note

LOGISTIC COST PRE-ANALYSIS COST TOTAL

SERVICE PARTNER DEBIT NOTE 15,00 € 34,00 € 49,00 €

Table 2 - Delphi Analysis and Handling Costs

LOGISTIC COST ANALYSIS COST HANDLING COST TOTAL

DELPHI ANALYSIS AND

HANDLING COSTS

15,00 € 33,00 € 60,00 € 108,00 €

Regarding the 153 Warranty Claims confirmed as supplier liability, there are 5 different defective

components that were analyzed, divided in 29 different part numbers, which are presented in the Table

3. The combination of the different components result in 56 radio models. Those radio have different

warranty costs associated, that varies according to the model of the radio.

Table 3 - Defective Components and Radios Part Numbers

COMPONENTS COMPONENTS P/N QUANTITY RADIO MODELS

BLUETOOTH MODULE 3

56

CD PLAYER 5

CIRCUIT BOARD 16

IC 2

TRANSISTOR 3

3.4.2 Database with Customer Debit Costs per Device


53

In the beginning of the project, the way to quantify all the costs for each warranty claim was to be

associated to the component and to the root cause. However, it was realized later that this was

impossible, since the cost does not depends on the root causes neither on the components, depends

only on the radio model. Since the Customer Debit Notes have a Technical Factor, that varies from

customer to customer, depending on variables as the complexity of the radio analysis, radio purchasing

price and dealers logistics, handling costs to collect the radio in the reference markets and on a estimative

done for the defective radios for the rest of the world, where the defective radios are not collected.

According to Customer Terms and Conditions, the Technical Factor is confidential, although Delphi should

know the variables and the fractions that integrate the formula, since the debited values varies even in

radios with the same Part Number. In order to surpass that situation, it was decided by Delphi that the

highest Customer Debited Value per device P/N would be the reference value for the current and future

Cost Recoveries. That decision resulted from the fact that every Customer Debit Note made to Delphi

does not have the DPRTS of each Warranty Claim, being impossible to link a specific warranty claim to a

Customer Debit Note. Is even more difficult, when the Customer Debit Notes are organized by Customer

Part Number and not by Delphi Part Number regarding each defective radio, creating the need for a

Database with all the devices P/N that is in place in Delphi from now on, organized by Customer P/N,

Delphi P/N, Delphi Selling Price, Device Designation, Customer and Customer Debit Note (Figure 49).

This database establish the Customer Debit Cost for each device that can appear in any Warranty Claim,

surpassing the previous uncertainty set by the Customer Technical Factor.

Figure 49 – Database Extract with Customer Debit Costs per Device


54

3.4.3 Warranty Claims Costs

All the warranty costs by defective component related to the 153 cases from supplier liability are

presented in the Table 4. The table shows the quantity of defective components divided according to the

analysis responsible, which can be only the service partner or Delphi and the Service Partner. If the

complaint root cause was already known, the complaint was managed by the service partner with a cost

of 49 euros, however when the root cause was new, the process followed to Delphi laboratory for further

analysis, adding 108 euros to the initial value, that sums up to a total of 157 euros per complaint analysis

cost. The analysis costs are summed to the Customer Debit Notes done to Delphi, what represents the

values that Delphi needs to debit to the responsible suppliers to recover the warranty costs incurred with

the 153 claims since 2012 until the First Quarter of 2015. The total amount that Delphi should recover

is 80.724.31€. The formal message presented to each responsible Supplier is showed at the Appendix

V.

Table 4 - Warranty Costs by Defective Component

Component Quantity Service

Partner

Delphi Pre-Analysis

Debit Note

Braga Analysis and

Handling Costs

Analysis Costs Customer Debit

Notes

Delphi Debit

Notes

Circuit Board 92 42 50

49,00 €

108,00 €

9.908,00 € 28.736,04 € 38.644,04 €

CD Player 32 25 7 2.324,00 € 21.556,13 € 23.880,13 €

BT Module 14 1 13 2.090,00 € 5.893,37 € 7.983,37 €

Transistor 10 1 9 1.462,00 € 4.969,44 € 6.431,44 €

IC 5 1 4 677,00 € 3.108,33 € 3.785,33 €

Total 153 70 83 16.461,00 € 64.263,31 € 80.724,31 €

As exhibited in the Table 4, the warranty costs can be displayed per defective component, however, it is

important to perceive them as debit notes that Delphi aims to present to the suppliers concerned. The

following Pareto Diagram in the Figure 50 shows the debit notes per supplied component, although with

a slight difference from the previous table. Due to the impossibility to use the name of each supplier, the

only component that is provided by two suppliers was nominated as Circuit Board A and as Circuit Board

B, with costs evaluated in 27.660,18 € of 63 supplied units and in 10.983,87€ of 29 units, respectively.

According to the debit notes values, the suppliers should refund Delphi in 80.724,31 €, distributing 77%

of the total cost by the companies that supplied the Circuit Board A, the CD player and the Circuit Board

B. The suppliers that provided the Bluetooth Module, the Transistor and the IC are responsible by the

other 23% of the total cost.


55

Figure 50 - The Debit Notes Distribution

When the perspective changes to the 24 root causes responsible for the 153 complaints presented in the

Table 5, the focus is on the Via Hole Open Issue and on the Catastrophic Optical Damage (COD). As

referred before, these root causes are accountable for more than half of the warranty claims and it seems

for more than half of the total cost with losses of 26.078,79 € and of 16.796,13 €, respectively.

Through the board observation, is possible to notice that despite the Circuit Board be the component with

most defects, the CD player is the component with most different root causes, which might indicate that

the number of different root causes is linked to the complexity of the component and to the number of

subcomponents that are assembled. The more number and variety of subcomponents it gets, the more

number of probable root causes appears.

€27.660,18

€23.880,13

€10.983,87

€7.983,37

€6.431,44

€3.785,33

34%

64%

77%

87%

95%

100%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

€-

€5.000,00

€10.000,00

€15.000,00

€20.000,00

€25.000,00

€30.000,00

Circuit Board A CD Player Circuit Board B Bluetooh Module Transistor IC

The Debit Notes Distribution

Debit Notes Comulative %


56

Table 5 - Delphi Debit Notes per Root Causes

Component Root Cause Number of Cases Delphi Debit Notes

CD Player

Catastrophic Optical Damage (COD) 22 16.796,13 €

Defective PCB 3 2.438,40 €

Deformed Roller 2 1.343,37 €

Defective OPU 1 732,90 €

Eject Lever Spring Bad Assembled 1 623,10 €

Gear Rack Spring Bad Assembled 1 700,03 €

Main Slider Deformed 1 623,10 €

Wafer Fabrication Defect 1 623,10 €

Bluetooth

Module

Wrong Pffs Auto detection 5 2.144,58 €

ASIC P5+ IC Defective 4 2.745,68 €

Internal Structure Impaired 2 1.071,53 €

BT Chip Defective 1 445,04 €

C116 Defective 1 766,88 €

U1702 IC Defective 1 809,66 €

Circuit Board

Via Hole Open Issue 60 26.078,79 €

Containers Contamination 13 3.146,67 €

Electrical Spikes 7 1.944,43 €

Crack at the Bottom of Laser Via 6 4.279,56 €

Solder Mask Surface Scratched 5 2.966,09 €

U6 IC Program 1 228,50 €

Transistor Package Corrosive Elements 9 5.708,32 €

Wire bonded Lead frame Contact 1 723,12 €

IC Random FAB Particle 4 2.973,72 €

Disturbed Vcc Bond Wires 1 811,61 €

3.4.4 Supplier Perspective

Finally, it is also important to get to know the supplier perspective regarding the Cost recovery process.

In the Figure 51 is shown the average price that Delphi pays for each type of supplied component in

comparison with the average cost per defective radio that Delphi should debit to each supplier. From the


57

supplier point of view, is hard to understand the fact that one single transistor, out of thousands, that is

sold in average to Delphi for 0.19 €, can damage a radio which average cost is charged to supplier for

643,14 €. That is why, it is very important to collect all the data as Delphi Analysis Reports, Supplier

Analysis Reports, Customer Debit Notes, Complaint Analysis and Handling Costs to convince the supplier

that the costs are real and must share the risk with Delphi when the liability is on his side. Having in

consideration Delphi Zero Defects Policy, all the supplied materials should arrive with no damage to the

customer facilities, as Delphi is committed to do with its customers. The supplier should always ensure

the quality of the product, whether it is through the implementation of corrective actions and contention

measures, whether it is by sharing the risk in warranty claims with Delphi.

Figure 51 - Supplier Perspective

As already mentioned, every company wants to avoid Costs, especially due to Poor Quality, regarding

Warranty Claims, which includes both Internal and External Failure Costs as the scrap, the analysis, the

handling and logistics, the dispersed information in several databases and the customer debit notes,

customer satisfaction, the relationship with suppliers, respectively. These costs can be only avoided, if

everyone is committed throughout the supply chain to prevent defects by learning from previous

experiences, through benchmarking with other Delphi factories, satisfying Customer Requirements. The

higher is the short-term investments in Costs of Good Quality as Prevention and Appraisal Costs, less are

the long-term costs of Poor Quality throughout the Supply Chain as Internal and External Failure Costs,

ensuring a win-win strategy for all stakeholders and nurturing the business partnership.

B L U E T O O T H M O D U L C D P L A Y E R

C I R C U I T B O A R D I C

T R A N S I S T O R

€13,62 €12,50

€3,24 €3,39

€0,19

€570,24 €746,25

€420,04

€757,07

€643,14

SUPPLIER PERSPECTIVE

Component Price Average Debit Note Average


58


59

4. CONCLUSIONS

In this chapter, it is presented the Final Considerations regarding the project development, the Limitations

that appeared along the way and the Future Work that could be done to improve the current results to

increase the Quality in the company daily routine.

4.1 Final Considerations

This dissertation was developed in the automotive industry in Delphi Automotive Systems Portugal S.A.,

in Braga, with the main objective to analyze the current warranty flow process and warranty claims since

2012 until the First Quarter of 2015, in order to implement a new way to recover the Quality Costs

regarding supplied defective material.

Currently, the quality costs regarding each warranty claim are unknown by Delphi, setting a fixed value

of 1000 € for every warranty claim, a figure that is not valid at suppliers eyes, since does not reflect the

real costs of each claim and it is much higher than the price that Delphi paid for each component.

Although, Delphi has a Zero Defects Policy and states that did not pay for No Quality regarding defective

components which are being charged by Customer. Delphi Braga wants to be the first company in Delphi

group to recover quality costs from every supplier, in order to raise awareness and to ensure that every

supplier is committed to Delphi Policy and to minimize the current quality costs.

When compared to Line Pull and Customer Impact, the Warranty Costs Recovery are higher since it

represents only those two suppliers under agreement with Delphi due to epidemic situations, setting aside

the suppliers with less complaints individually, who are the object of study for this project. Even though,

there are great differences between the amount proposed by Delphi and what is the value accepted by

the suppliers under agreement.

According to the literature, one of the principles of quality continuous improvement is based on the

assumption that every decision, in special those which are taken by the quality team should rely on the

use of Quality Tools. That is why the Quality Tools are used to improve the Quality Costs, for the purpose

to discover the critical root causes to be solved and to implement corrective actions, ensuring the Quality

Standards in the beginning of the Supply Chain. The higher is the short-term investment in Costs of Good

Quality as Prevention and Appraisal Costs, the less are the long-term costs of Poor Quality throughout the

Supply Chain as Internal and External Failure Costs, ensuring a win-win strategy for all stakeholders and

nurturing business partnership towards Delphi requirements of being known among customers as their

best supplier, surpassing their highest expectations.


60

In order to discover how many complaints were caused by defective material, a deep analysis was

undertaken to the 7696 complaints from 2012 until the First Quarter of 2015, which 6822 of the cases

are closed and 874 are ongoing, representing 89% and 11% of the pie chart distribution, respectively.

The Pareto diagram identified that the most number of complaints is concentrated in 3 customers that

represents 83% of the total cases, out of 13 customers. Taking into consideration the analysis for defects

liability, customer is responsible for 57%, supplier for 34% and Delphi for 9% of the closed complaints.

Since the focus of the analysis is on the complaints caused by supplier, the 2297 concerned cases was

scrutinized, although due to DPRTS database lack of information, there was the need to create the

Supplier ID and Component Part Number in order to identify the supplier and the respective defective

component. Towards that purpose, it was possible to associate the respective supplier to 1961 of the

cases, although the component part number is only identified in 1415 of those cases. If there was a

database that centered all the needed piece of information, the information flow would be improved

significantly, since it would allow to look across the 3 type of problems. The same root cause can affect

the same component P/N or the same device P/N across the 3 types of problems, that is why it is

important to see the big picture and to discover similarities between problems to have a concerted action

in their solution and to prevent Costs of Poor Quality.

In order evaluate the Quality Indicators, the Scatter Diagram was used to identify if there is a relation

between the period of warranty and the car mileage regarding the 1415 cases. This analysis concluded

that 45 cases were out of warranty, what means that the majority of the complaints occurred within the

car warranty period of 730 days, proven that the warranty period is enough to guarantee final customers

assistance. Regarding the car mileage record with an average of 16669 km, when compared to the

highest value of the distribution with 226982 km, the value is relatively low, leading to the evidence that

the mileage average when defects occurs is relatively early. When the focus is on the cases out of warranty

and on the cases with a mileage above the average, a total of 22 cases, is noticed that there is not a

direct relation between the car mileage and the warranty period.

The time that takes to conclude a warranty claim and a complaint analysis is a quality indicator that is

evaluated by the customer and matters to customer satisfaction. In Braga the analysis duration takes in

average 6 days, which is half of the 12 days that the services partners take to analyze each complaint.

The results are surprising, since Delphi Braga has a lack of laboratory analysts, having in consideration

that 3 people analyses all the defective devices send by the customer, whether are warranty or 0 km

issues, although it represents only 5% of the total warranty claims that Delphi Braga is responsible for.

Due to logistics and handling issues, Delphi relies on outsourcing Service Partners that collect and analyze


61

the defective radios majority. On the other hand, Delphi Braga takes 21 days in average to process a

complaint, in contrast with Service Partners, who spends 37 days in average to close a warranty claim.

Since, all the complaints from new projects will be analyzed in Braga, these numbers are a good indication

of what should be expected, although there will be a service increasing, what will demand an increasing

of the available capacity, in order to maintain or increase the customer satisfaction.

Afterwards, it is possible to observe that supplier analysis rejected the liability in 66% of the cases,

approving 20% of the cases and 14% are inconclusive, due to the lack of supplier 8D Report in each case.

The 20% that were approved by supplier analysis corresponds to 280 cases, divided by 153 that are

supported by supplier analysis 8D Report, 93 cases that are caused by the couple of suppliers with whom

Delphi has an agreement with and the remain 34 cases are from different suppliers, representing less

than 5 cases per supplier, which are postponed until there are new warranty complains regarding those

suppliers, in order to increase the suppliers receptivity.

After a rigorous analysis, from the 7696 complaint cases, there are 153 complaints confirmed by supplier

analysis as defective material, where was possible to associate every quality cost to each warranty claim.

If the complaint is managed entirely by the service partners, the costs associated are separated in 15

euros for the Logistic Costs and in 34 euros for the Pre-Analysis Cost, with a total of 49 euros. In case of

the pre-analysis results concludes that the root cause of the problem has never seen before, the radio is

send to Delphi Braga for further analysis. The costs associated to Delphi Braga analysis consists in 15

euros for the Logistic Costs, in 33 euros for the analysis costs and in 60 euros for handling costs that

sums up in 108 euros plus the 49 euros from the Service Partner Debit Note, with a total of 157 euros.

Regarding the 153 Warranty Claims confirmed as supplier liability, there are 5 different defective

components that were analyzed, divided in 29 different part numbers. The combination of the different

components result in 56 radio models. Those radio have different warranty costs associated, that varies

according to the model of the radio. Therefore, a Database was created with all the devices P/N that is

in place in Delphi from now on, organized by Customer P/N, Delphi P/N, Delphi Selling Price, Device

Designation, Customer and Customer Debit Note. This database establish the Customer Debit Cost for

each device that can appear in any Warranty Claim, surpassing the previous uncertainty set by the

Customer Technical Factor and by Customer Debit Notes lack of DPRTS ID. The analysis costs are

summed to the Customer Debit Notes done to Delphi, what represents the values that Delphi needs to

debit to the responsible suppliers to recover the warranty costs incurred with the 153 claims since 2012

until the First Quarter of 2015. The total amount that Delphi should recover is 80.724.31€.


62

Through the resource to Pareto diagram is presented that Circuit Board and CD Player are the most

defective components affected by complaints, representing 81% of the cases, followed by Bluetooth

Modules, Transistors and IC.

Since all the defective components in analysis are electronic components, it is possible to conclude that

these parts were supplied already pre-damaged due to sensibility factors as electrical discharges, which

failure modes only appears under certain conditions of stress such as impact, heat, humidity and dust

that wear out the component and that are only felt when the radio is already on the road. Unfortunately,

this kind of situations are hard to detect before the radio arrival to the customer hands, that is why the

company should share this risks and costs with the supplier that supplied the defective component and

improve the process continuously to prevent this kind of defects.

The 5 types of defective components were supplied by 6 Suppliers, being responsible by each type of

component, excepting the circuit boards, the component with most complaints, provided by 2 suppliers,

with 63 and 29 of the complaints. According to the debit notes values, the suppliers should refund Delphi

in 80.724,31 €, distributing 77% of the total cost by the companies that supply circuit boards and cd

players. The suppliers that provided the Bluetooth Module, the Transistor and the IC are responsible by

the other 23% of the total cost.

Due to complexity and extension of the information presented in DPRTS, there was the need to create

Root Cause Categories, which are used by Delphi from now on in future claims, regarding the 24 root

causes linked to the 153 confirmed complaints. Via Hole Open Issue and Catastrophic Optical Damage

(COD) are the top root causes in 60 and 22 of the cases under scope, representing more than half of the

total cases and are accountable also for more than half of the total quality costs of 26.078,79 € and of

16.796,13 €, respectively. Since these problems affected several auto radios assembled with supplied

components from different models and from different production batches. Despite the Circuit Board be

the component with most defects, the CD player is the component with most different root causes, which

can be concluded that the number of different root causes is linked to the complexity of the component

and to the number of subcomponents that are assembled. The more number and variety of

subcomponents it gets, the more number of probable root causes appear. That is why, it was mandatory

to discover the origin of these root causes and the corrective actions to implement, through the resource

of Quality Tools as the 5Whys method and the Cause and Effect Diagram by Ishikawa, aided by Focus

Group methodology, resulting in correctives actions that were put in place by supplier, to ensure the

quality in the beginning of the supply chain. Regarding the Via Hole Open Issue, the 5 Whys allowed to


63

put in place corrective actions as boards surface cleaning, an alarm that is triggered every time there is

a Broken Drill and the Work Instructions improvement to facilitate the job of the operator. When come to

the Optical Damage, the Cause and Effect Diagram allowed the problem solving through the application

of Electrostatic Discharge Bags to protect the Cd Players and a mechanism to limit the current escalation.

From the supplier point of view, is hard to understand the fact that one single transistor, out of thousands,

that is sold in average to Delphi for 0.19 €, can damage a radio which average cost is charged to supplier

for 643,14 €. That is why, is very important to collect all the data that is important, as Delphi Analysis

Reports, Supplier Analysis Reports, Customer Debit Notes, Complaint Analysis and Handling Costs to

convince the supplier about the validity of the values to proceed to Costs Recovery.

According to information from Supplier Quality responsible, this new methodology to recover quality costs

is going to be used for the future warranty claims, having into consideration supplier’s feedback. Since

this is a new system that was never used before, Delphi Braga will benchmark this methodology to other

companies of the group. This project contributed to a personal development, allowing learning

opportunities that will enrich and endure throughout the Future.

4.2 Limitations

During the project there were some limitations that created some difficulties to accomplish the proposed

objectives, which were confidential issues regarding the Technical Factor present in every Customer Debit

Note and the access to Delphi Terms and Conditions established with each Supplier. On the other hand,

the fact that all the valuable information concerning Projects Policy from Delphi Braga is centered in

Delphi Germany, was an issue of struggle.

Even though, there were some limitations, the proposed objectives were accomplished with success.

4.3 Future Work

Regarding the future work and the continuous improvement of the current process, it is recommended to

Delphi to:

1. Evaluate the acceptance rate of supplier analysis regarding defective components liability after the

costs recovery implementation and compare with the present scenario;

2. Evaluate the currency devaluation that Delphi has regarding the purchase of raw materials in dollars

to sell the final materials in euros;


64

3. Benchmark with other Delphi factories to replicate the cost recovery practices and to standardize the

process;

4. Following the cost recovery implementation and improve the process from supplier feedback as an

agreement negotiation in the beginning of each project that set a fixed value to share the warranty risks,

improving the relation with suppliers and reducing the pressure in each complaint analysis;

5. Concentrate all the complaints data regarding the different type of defects in one data base that allowed

the look across between cases, to study trends and correlations, for a robust concerted action regarding

corrective measures;

6. Implementation of the warranty cost recovery in a monthly basis with the Customer Debit Notes

corresponding to each specific case.


65

BIBLIOGRAPHY

Anderson, E. W., Fornell, C., & Rust, R. T. (1997). Customer Satisfaction, Productivity, and Profitability:

Differences Between Goods and Services. Marketing Science, 16(2), 129-145.

doi:10.1287/mksc.16.2.129

Aniza, L., Wang, M. H., & Rieger, F. (2013). Development of Quality Cost Model within a Supply Chain

Environment. AMM Applied Mechanics and Materials, 330, 737-742.

Arter, D. R. (2003). Quality audits for improved performance.

Ayaǧ, Z., & Samanlioglu, F. (2014). An intelligent approach to supplier evaluation in automotive sector.

Journal of Intelligent Manufacturing.

Battaglia, D., Schimith, C. D., Marciano, M. A., Bittencourt, S. A. M., Diesel, L., Borchardt, M., & Pereira,

G. M. (2014). Creating Value through Services and Relationships: The Perception of Purchasing

Companies. Procedia CIRP, 16, 26-31. doi:http://dx.doi.org/10.1016/j.procir.2014.01.004

Bolfing, C. (1989). How do Customers Express Dissatisfaction and What can Service Marketers do About

it? Journal of Services Marketing, 3(2), 5-23. doi:10.1108/EUM0000000002483

Bosch, V. G., & Enríquez, F. T. (2005). TQM and QFD: exploiting a customer complaint management

system. International Journal of Quality & Reliability Management, 22(1), 30-37.

doi:10.1108/02656710510572977

Cabral, Colaço, & Guerreiro. (2002). A qualidade em Portugal. Lisboa.

Chen, H.-R., & Cheng, B.-W. (2010). A case study in solving customer complaints based on the 8Ds

method and Kano model. Journal of the Chinese Institute of Industrial Engineers, 27(5), 339-

350. doi:10.1080/10170669.2010.495508

Crosby, P. B. (1979). Quality is free : the art of making quality certain. New York: McGraw-Hill.

Gerson, R. F. (1998). Medir a qualidade e a satisfação do cliente: um guia para gerir um serviço de

qualidade. Lisboa: Monitor.

Gilmore, H. L. (1970). The cost of product conformance quality control: a research study into

conformance quality control cost of selected consumer product manufacturing companies. By

Harold Lawrence Gilmore. [Syracuse, N.Y. Available from http://worldcat.org /z-wcorg/

database.

http://dx.doi.org/10.1016/j.procir.2014.01.004

http://worldcat.org/


66

Hagemeyer, C., Gershenson, J., & Johnson, D. (2006). Classification and application of problem solving

quality tools: A manufacturing case study. The TQM Magazine, 18(5), 455-483.

doi:10.1108/09544780610685458

Islam, K. A. (2006). Developing and measuring training the six sigma way a business approach to training

and development.

Juran, J. M., & De Feo, J. A. (2010). Juran's quality handbook : the complete guide to performance

excellence. New York: McGraw Hill.

Juran, J. M., & Gryna, F. M. (1988). Juran's quality control handbook. New York: McGraw-Hill.

Kathuria, R., & Davis, E. B. (1999). Quality and Work Force Management: From Manufacturing Managers'

Perspective. Journal of Quality Management, 4(2), 147-166. doi:10.1016/S1084-

8568(99)00010-3

Krishnan, S. K., Agus, A., & Husain, N. (2000). Cost of quality: The hidden costs. Total Quality

Management, 11(4-6), 844-848. doi:10.1080/09544120050008309

Manal, Y., Joo, J., & Shouming, C. (2013). TQM, strategy, and performance: a firm‐level analysis.

International Journal of Quality & Reliability Management, 30(6), 690-714.

doi:10.1108/02656711311325638

McAlister, D. T., & Erffmeyer, R. C. (2003). A content analysis of outcomes and responsibilities for

consumer complaints to third-party organizations. Journal of Business Research, 56(4), 341-351.

doi:http://dx.doi.org/10.1016/S0148-2963(02)00445-9

Montgomery, D. C. (2009). Introduction to statistical quality control. Hoboken, N.J.: Wiley.

Morgan, D. L. (1997). Focus groups as qualitative research. Thousand Oaks, Calif.: Sage Publications.

Muffatto, M., & Panizzolo, R. (1995). A process‐based view for customer satisfaction. International

Journal of Quality & Reliability Management, 12(9), 154-169.

doi:10.1108/02656719510101259

Oakland, J. S. (1995). Total quality management : the route to improving performance. Oxford; Boston:

Butterworth-Heinemann.

O´Brien, R. (1998). An Overview of the Methodological Approach of Action Research: Faculty

of Information Studies: University of Toronto.

Paliska, G., Pavletic, D., & Sokovic, M. (2007). Quality tools : systematic use in process industry. Journal

of achievements in materials and manufacturing engineering(26), 79-86.

http://dx.doi.org/10.1016/S0148-2963(02)00445-9


67

Piskar, F., & Dolinsek, S. (2006). Implementation of the ISO 9001: from QMS to business model.

Industrial Management & Data Systems, 106(9), 1333-1343.

doi:10.1108/02635570610712609

Pyon, C. U., Woo, J. Y., & Park, S. C. (2011). Service improvement by business process management

using customer complaints in financial service industry. Expert Systems with Applications, 38(4),

3267-3279. doi:http://dx.doi.org/10.1016/j.eswa.2010.08.112

Reed, R., Lemak, D. J., & Mero, N. P. (2000). Total quality management and sustainable competitive

advantage. Journal of Quality Management, 5(1), 5-26. doi:http://dx.doi.org/10.1016/S1084-

8568(00)00010-9

Sampaio, P., Saraiva, P., & Rodrigues, A. G. (2011). The economic impact of quality management

systems in Portuguese certified companies: Empirical evidence. International Journal of Quality

& Reliability Management, 28(9), 929-950. doi:10.1108/02656711111172522

Schiffauerova, A., & Thomson, V. (2006). A review of research on cost of quality models and best

practices. International Journal of Quality & Reliability Management, 23(6), 647-669.

doi:10.1108/02656710610672470

Sousa, S. D., Aspinwall, E., Sampaio, P. A., & Rodrigues, A. G. (2005). Performance measures and quality

tools in Portuguese small and medium enterprises: survey results. Total Quality Management &

Business Excellence, 16(2), 277-307. doi:10.1080/14783360500054434

Talay, M. B., Calantone, R. J., & Voorhees, C. M. (2014). Coevolutionary Dynamics of Automotive

Competition: Product Innovation, Change, and Marketplace Survival. Journal of Product

Innovation Management, 31(1), 61-78. doi:10.1111/jpim.12080

Tarı́, J. J., & Sabater, V. (2004). Quality tools and techniques: Are they necessary for quality management?

International Journal of Production Economics, 92(3), 267-280.

doi:http://dx.doi.org/10.1016/j.ijpe.2003.10.018

Tereso, A. (2011). Slides das Aulas da Unidade Curricular de Metodologias de Investigação. Guimarães,

Portugal: Departamento de Produção e Sistemas.

Tsai, W. H. (1998). Quality cost measurement under activity-based costing. International Journal of

Quality and Reliability Management, 15(7), 719-752.

Tsim, Y. C., Yeung, V. W. S., & Edgar, T. C. L. (2002). An adaptation to ISO 9001:2000 for certified

organisations. Managerial Auditing Journal, 17(5), 245-250.

doi:10.1108/02686900210429669

http://dx.doi.org/10.1016/j.eswa.2010.08.112

http://dx.doi.org/10.1016/S1084-8568(00)00010-9

http://dx.doi.org/10.1016/S1084-8568(00)00010-9

http://dx.doi.org/10.1016/j.ijpe.2003.10.018


68

Tzelepis, D., Tsekouras, K., Skuras, D., & Dimara, E. (2006). The effects of ISO 9001 on firms' productive

efficiency. International Journal of Operations & Production Management, 26(10), 1146-1165.

doi:10.1108/01443570610691111

Wahid, R., & Corner, J. (2009). Critical success factors and problems in ISO 9000 maintenance.

International Journal of Quality & Reliability Management, 26(9), 881-893.

doi:10.1108/02656710910995073

Wellburn, J. (1996). A TQM life cycle case study. The TQM Magazine, 8(3), 35-45.

doi:10.1108/09544789610118467

Zeithaml, V. A., & Bitner, M. J. (2000). Services marketing : integrating customer focus across the firm.

Boston: Irwin/McGraw-Hill.


69

APPENDIX I – COST RECOVERY PROCEDURE

Figure 52 - Cost Recovery Procedure


70


71

APPENDIX II– RADIO COMPONENTS

Figure 53 - Radio Components


72


73

APPENDIX III - WARRANTY FOLLOW-UP INSTRUCTIONS

Figure 54 - Warranty Follow-up Instructions


74


75

APPENDIX IV – ROOT CAUSE CATEGORIES

Table 6 - Root Cause Categories

Root Causes Categories

A I

Air Buble Inside IC U55 Defective

ASIC P5+ IC Defective IC1601 Output Driver

B Intermittent Image

Bad Assembled Internal Structure Impaired

Broke Seek Button Backward Snap L

Broke Seek Button Forward Snap Leakage Path by Tungsten Residue

BT Chip Defective LED 5 Open Circuit

C Loose Clamp

C116 Defective M

Catastrophic Optical Damage (COD) Main Slider Deformed

Catastrophic Optical Damage (COD) Missing Component

Containers Contamination Missing Segments

Crack at the Bottom of Laser Via P

Current Leaks Package Corrosive Elements and Moisture

D R

Defective Pin 4 SD_VDD Random FAB Particle

Defective Antenna Signal Reception Residue Contamination

Defective CIC U12 S

Defective GPS Tuner Soft Bridge Short

Defective IC U20 MOS Solder Crack On OPU

Defective IC U48 Solder Mask Surface Scratched

Defective LED solder pads T

Defective OPU Test Deficiency Issue

Defective PCB TNF

Defective Wiper Turn Tracking Coil Disconnected

Deformed Roller U

Display White Stain U1702 IC Defective

Disturbed Vcc Bond Wires U6 IC Progamm

Drive Failure V

E Via Hole Open Issue

Eject Lever Spring Bad Assembled W

Eject Lever Spring Failure Wafer Fabrication Defect

Electrical Connection Wirebonded Leadframe Contact

Electrical Spikes Wrong Pffs Autodetection

EOS

External Object Inside

G

Gear Rack Spring Bad Assembled

GPS Tunner Defective


76


77

APPENDIX V – COST RECOVERY FORMAL MESSAGE

Hello _____ Team,

On behalf of Delphi Braga Plant and in name of Supplier Quality Engineering, we are contacting you

regarding some problem cases that took place in Warranty. In total, we are talking about ________ that

reached the Final Customer and failed on the road, which consequence was the automotive dealers call

to disassemble the ________ from the collected vehicles and their replacement. The defective _______

were gathered to be analyzed the root cause behind each defect.

As you know, each case is unique, since a failure mode can occur due to different root causes and vice

versa. Having that in mind and according to Delphi Analysis, confirmed by ______8D Reports, in every

single case, the failure mode took place due to a defective component.

The _________ were analyzed by _______, and the analysis result concluded the issue was ________

liability.

Our End Customer charged Delphi for each Warranty Claim.

Therefore, Delphi wants to be reimbursed for the costs charged so far, and thus wants to impute those

costs to _________, who assumed the liability, since the root cause source behind every issue is in your

production system.

As you are aware, Delphi focus is Zero Defects Policy, since we are fully committed to exceed our

Customer´s Greatest Expectations being their Best Supplier. Its Delphi’s duty, to assure that these values

are followed by every Supplier to guarantee the Quality of the Process in every step of the way, to meet

Customer Requirements. According to our values, every piece of material supplied to Delphi should be in

conformity before enter in our plant and when this criteria is not fulfilled, the responsible one should

assume it’s liability.

Like in every Successful Partnership, only when we are on the same page, sharing the same values led

by continuous improvement, we can achieve Success and guaranty a Sustainable Future.

We appreciate your comprehension and notice that our Success is your Success!

Best Regards,

Delphi Braga Plant SQE Team

Documents

André Luís Rocha da Silva Costarepositorium.sdum.uminho.pt/bitstream/1822/39287/1/André... · 2019. 1. 1. · Universidade do Minho Escola de Engenharia André Luís Rocha da Silva