FUZZY QUALITY FUNCTION DEPLOYMENT (FUZZY-QFD) …

International Association for Management of Technology IAMOT 2018 Conference Proceedings

FUZZY QUALITY FUNCTION DEPLOYMENT (FUZZY-QFD) APPLIED TONEW DEFENSE PRODUCT DEVELOPMENT

MARIA FATIMA LUDOVICO DE ALMEIDAPontifical Catholic University of Rio de Janeiro, MSc. Program on Metrology, Brazil

[email protected] (Corresponding)

CARLOS EDUARDO SILVA DA LUZArmy Technology Center (CTEx), Brazil

[email protected]

GUILHERME DE ANDRADE MARTINS Pontifical Catholic University of Rio de Janeiro, MSc. Program on Metrology, Brazil

[email protected]

ABSTRACT

The Quality Function Deployment (QFD) tool integrated with multicriteriadecision-support methods has been widely applied to new product design,particularly with the support of fuzzy logic. The objective of this paper is topropose a conceptual model based on an integrated fuzzy-ANP-QFD approach todetermine and prioritize engineering requirements of new defense products inlight of future users’ perspective. The research can be considered descriptive,applied, and methodological. Based on the results of the bibliographic anddocumentary review on the central themes of the research, a conceptual modelwas developed to determine and prioritize engineering requirements of newdefense products, seeking to fill gaps identified during the literature reviewcovering the period of 1987-2017. The paper emphasizes the opportunity toexplore an integrated fuzzy-ANP-QFD approach in projects of new defenseproducts since applications in this area were not identified during the literaturereview. The applicability of the model was demonstrated by an empirical casestudy having as experimental context the COBRA 2020 Project, a strategicinitiative of the Brazilian Army. For this study, one of the new products to bedeveloped within this Project was selected – a new thermal vision monocular.This work provides a conceptual model based on an integrated fuzzy-ANP-QFDapproach for academicians and researchers to determine and prioritizeengineering requirements of new defense products in a more efficient waycompared with current practices. The research findings have the potential to bereplicated in other projects of new defense products - at the Army TechnologyCenter (CTEx) - and other military institutions dealing with research,development, and innovation (RD&I) activities in Brazil and abroad.

Key words: Quality Function Deployment; Fuzzy-ANP-QFD; New product design;Defense; COBRA 2020 Project.

INTRODUCTION

Page 1 of 20


The development of new defense products to meet big challenges faced bycombatants in their various modalities reinforces the importance of proposing aconceptual model that effectively aligns the design requirements of theseproducts to the needs and expectations of future users.

To include the perspective of future users of new defense products can be adifferential over current practices in military contexts all over the world. Most ofthe projects of defense products do not consider the dual-use of theseproducts/technologies into their scope, given their unique specificities alignedwith Army doctrine and operational guidelines. Despite the existence of manyexamples of dual-use products designed for both military and civilian purposes,in general, military R&D teams have been prioritizing engineering requirementsof new defense products based almost exclusively on criteria and parametersdefined in technical standards and normative guidelines, without considering theexpectations and needs of military users (combatants).

In this context, the purpose of this work is to present a conceptual model basedon an integrated fuzzy-ANP-QFD approach to determine and prioritizeengineering requirements of new defense products in light of future users’perspective. The applicability of the model could be demonstrated by anempirical case study having as experimental context the COBRA 2020 Project, astrategic initiative of the Brazilian Army. For this study, one of the new productsto be developed within this project was selected – A new thermal visionmonocular.

In a new product design process, concept selection is a very important issue dueto its strong influence in both upstream and downstream activities of this area.Consequently, several methods have been introduced to concept selection, beingthe most cited Analytical Hierarchy Process (AHP); Quality Function Deployment(QFD) tool; Analytical Network Process (ANP); fuzzy QFD; and combination of QFDwith multicriteria decision-making methods (MCDM) and fuzzy logic.

The Quality Function Deployment (QFD) tool combined with MCDM methods andfuzzy logic has been extensively applied to the design of new products in a widerange of sectors (Maritan, 2015; Younesi and Roghanian , 2015; Chen et al.,2015; Zaim et al., 2014; Abdolshah and Moradi, 2013; Day and Blackhurst, 2012;Weng et al., 2009; Lee et al., 2008; Zhou, 1998). In spite of the increasedworldwide interest in applications of fuzzy-MCDM-QFD in new product design, nowork focusing on new defense products was identified in the period from 1987 to2017. These findings reinforced the opportunity to explore an integrated fuzzy-ANP-QFD approach in new defense product design.

The article is structured as follows: firstly, we conceptualize Quality FunctionDeployment (QFD) and briefly review the approaches and models for determiningand prioritizing engineering requirements of new products in light of future users’perceptions. Based on a literature review on MCDM methods and fuzzy logic, weconfirmed the opportunity to investigate an integrated QFD-MCDM model appliedto new product design in a fuzzy environment. In section 3, we describe theresearch methodology in six stages. In Sections 4 and 5, we introduce the

Page 2 of 20


conceptual model and discuss the results of an empirical study focusing thedevelopment of a new defense product within the COBRA 2020 Project. Finally,in Section 6, we synthesize the concluding remarks and point out implications forfuture works.

THEORETICAL BACKGROUND

The theoretical background encompasses the following themes: (i) QualityFunction Deployment (QFD); and (ii) MCDM methods and fuzzy logic combined toQFD in new product design context.

Quality Function Deployment (QFD)

Quality Function Deployment (QFD) is defined as a tool that integrates customerneeds throughout the development cycle of a new product. QFD converts therequirements and expectations of future users into engineering features(engineering requirements) and transfers them to the subsequent stages of theproduct development through to manufacturing, through successive unfolding(King, 1987; Hauser and Clausing, 1988; Clausing, 1994; Akao and Mazur, 2003;Maritan, 2015).

There are different versions of QFD, the most well-known being the onecharacterized by four major phases, namely: (i) product planning; (ii)development; (iii) process planning; and (iv) production planning (Akao, 1990;Kahraman et al., 2006; Liu, 2009). To evaluate the interrelations between therequirements of each phase "Houses of Quality" (HoQ) are constructed. In fact,the central methodological element of the QFD tool is the "House of Quality"(Sharma et al., 2008; Kahraman et al., 2006).

Engineering and customer requirements should be measurable to behierarchical. Thus, the CRs versus ERs matrix is performed by calculating therelative weights for each of these requirements matrices. to minimize bias in thehierarchy of requirementS and maximize customer satisfaction, this workproposes to integrate a multicriteria method of decision support and fuzzy logicto the QFD tooL.

It should be emphasized that for purposes of the intended modeling, the focus ofapplication of QFD will only contemplate Phase I - Product Planning.

MCDM methods and fuzzy logic combined to QFD in new product design context

Several researchers initially used the Analytic Hierarchy Process (AHP) introducedby Saaty (1980), combined with QFD, to determine the degree of relativeimportance of the quality requirements of a new product from the perspective offuture users (Wang et al., 1998; Tu et al., 2010; Mayyas et al., 2011; Dai andBlackhurst, 2012; Chen et al., 2015).

Page 3 of 20


Due to the imprecision and uncertainty in the judgments regarding the degree ofimportance of the quality requirements of a new product, several authors beganto integrate multicriteria methods of decision support and fuzzy logic to the QFDtool.

In this Section, we present the results of the bibliographic and documentaryreview, covering the period from 1987 to 2017 and focusing on the QFDapproach in new product design, which combine multicriteria methods ofdecision support and fuzzy logic to QFD. To this end, keywords were defined asproduct development; product design; quality function deployment; QFD; MCDM;multi-criteria decision-making methods; fuzzy logic. The AND and OR operatorswere used to arrive at their respective search strategies in the Scopus, ScienceDirect and Web of Science databases. These strategies revealed the mostrelevant scientific works aligned to the objectives of the present research,considering the period 1987-2017.

Among the most relevant scientific studies, the combination of severalmulticriteria methods of decision support and fuzzy logic to QFD was evidenced,namely: (i) ANP (Raharjo, Brombacher and Xie, 2008); (ii) fuzzy Delphi and fuzzyDEMATEL (Wang, 2010); (iii) fuzzy DMS and fuzzy AHP (Güngör, Delice andKesen, 2011; Ho et al., 2012); (iv) fuzzy ANP (Kahraman et al., 2006).

We also observed that the QFD tool combined with these methods and fuzzylogic has been applied in several stages of Phase I (product planning), such as: (i)identification of customer priorities; (ii) prioritization of customer requirements;(iii) definition of design requirements; and (iv) prioritization of designrequirements in light of customer requirements.

The concept of fuzzy sets was first introduced by Zadeh (1965) to model theuncertainty in parameter definition, considering the subjectivity and experienceof the expert professionals. The use of fuzzy logic in decision-making processesallows us to convert inaccurate and described information into natural languagein numerical formats (Emrouznejad and Ho, 2017).

It was verified by comparing the methods and models identified during theliterature review that the integration of the ANP method into fuzzy-QFD modelswill allow to reveal dependency and feedback relationships between engineeringrequirements and customer requirements regarding new defense products.

RESEARCH METHODOLOGY

The research methodology framework comprised six stages, as shown in Table 1.

Table 1: Research methodology framework

Phase Stage Research questionSection

Motivation(Why?)

Problem definition.

Why should we develop the model?

Section 1

Conceptualization and development (What and

State of research on central themes and identification of research gaps and

Which MCDM methods can be used for determining and prioritizing engineering requirements of new defense

Section 2

Page 4 of 20


How?)

unsolved problems in the interface between the subfield of new defense product design and the MCDM-QFD?.

products in light of future users’ perceptions and needs?What are their limitations? And advantages in comparison with other methodological approaches?

Definition of the research methodology.

How could the model be developed and be demonstrated by an empirical study?

Section 3

Development of the integrated fuzzy-MCDM-QFD model.

Which methods should be combined to overcome limitations of current research on the new defense product design?

Section 4

Validation(How effectiveis the proposed model?)

Design of the empiricalstudy within the COBRA 2020 Project.

Among the new defense products contemplated in theCOBRA Project, which should be chosen for the empirical study?

Section 5

Empirical results and discussion of the managerial and policy implications of the model for Brazilian Army.

What are the methodological differentials of the proposed model compared to the current practices for prioritizing engineering requirements of new defenseproduct design?

Could the results of the empirical study demonstrate the effectiveness of the proposed model? Which are the managerial and policy implications of this research?

Several research methods were selected to address the questions posed in Table1.

Firstly, a literature review covered the main sources of peer-reviewed scientificarticles, such as Web of Science; Scopus; and Science Direct. Additionally,Google Scholar was accessed to complement the search results. the analysis ofcurrent state of research on central themes – new defense product design andfuzzy MCDM methods – led to the identification of research gaps concerning theinterface between these two important subfields.

Second, formal modeling was used to develop an integrated fuzzy-MCDM-QFDapproach for determining and prioritizing engineering requirements of newdefense products in light of future users’ perceptions.

Third, an empirical study focusing the development of a new thermal visionmonocular was carried out within the context of the COBRA 2020 Project. Thisstudy could integrate perceptions from both sides – R&D teams from the ArmyTechnology Center (CTEx) involved in this project and future customers(combatants) – to empirically validate the model, and to detect its limitationsand opportunities for improvement.

A FUZZY-ANP-QFD MODEL FOR NEW DEFENSE PRODUCT DESIGN

Page 5 of 20


This Section introduces the fuzzy-ANP-QFD model proposed for determining andprioritizing engineering requirements of new defense products in light of futureusers’ perceptions and needs.

During the process of prioritizing engineering or customer requirements in newdefense product design, uncertainties, inaccuracies of judgments or even amistaken consensus can arise from the articulation (or disarticulation) of theexperts and decision makers involved. To help prioritize these requirements inmilitary contexts, we adapted the model proposed by Kahraman et al. (2006),which integrates the Analytic Network Process (ANP) method and fuzzy logic tothe QFD. Figure 1 schematically represents the fuzzy-ANP-QFD approach appliedto new defense product design.

Figure 1: Fuzzy-ANP-QFD approach applied to new defense product design

The ANP method was proposed by Saaty in 1996, as an extension of the AHPmethod also created this author (Saaty, 1980). While the AHP decomposes aproblem at several levels, in such a way that they form a hierarchy, the ANPmethod can be used as an effective tool in cases where the interactions betweenthe elements of a system structure must be considered (Saaty, 1996).

Based on the theoretical framework presented in Section 2 and considering theArmy Doctrine and Operational Guidelines for the development of new defenseproducts, a conceptual modeling based on the fuzzy ANP-QFD approach isshown in Figure 2.

Page 6 of 20


Figure 2: Conceptual model based on the fuzzy-QFD approach for new defenseproduct design

Following, we describe the steps that integrate the conceptual model, according to the Figure 2.

Step 1. Definition of customer requirements (CRs) and association with engineering requirements (ERs)

From the analysis of the Army Doctrine and Operational Guidelines for thedevelopment of a given new defense product, the customer's requirements aredefined and grouped by direct consultation with future users of the focusedproduct. The engineering requirements (ERs) should be defined and alsogrouped by hearing members of the R&D teams involved in this development,considering applicable technical standards.

Once the client requirements (CRs) and engineering requirements (ERs) havebeen defined, the first stage of the construction of the House of Quality (HoQ) iscarried out.

Page 7 of 20


Step 2. Determination of the degrees of importance of CRs, with linguistic terms

In this step, the customer requirements (CRs) defined in step 1 are submitted tothe potential users’ opinions, in order to obtain the pairwise comparison matrixof the CRs, based on judgments about the degree of importance of each CR,using linguistic terms, with triangular fuzzy numbers, as presented in Table 2.

Table 2: Degrees of importance in fuzzy scales, according to Saaty (1996)

Linguistic scale fordetermining the

degree ofimportance

Fuzzy triangular scale Reciprocal scale

EIEqual importance

1 1 1 1 1 1

LI Low importance 1/2 1 1 1/2 2/3 1 2

MIModerate importance

1 1 1/2 2 1/2 2/3 1

SIStrong importance

1 1/2 2 2 1/2 2/5 1/2 2/3

VSIVery strong importance

2 2 1/2 3 1/3 2/5 1/2

AIAbsolute importance 2 1/2 3 3 1/2 2/7 1/3 2/5

By definition, the triangular fuzzy number (1, 1, 1) is used when two attributesare considered equally important (level of importance equal to 1 in the Saatyscale). With the pairwise comparison matrix, we get the eigenvector W1,calculated from the degree of importance of the customer requirements (CRs),using an expanded analysis of the fuzzy AHP method. Likewise, the relativeweights of W2, W3, and W4 are obtained (See Figure 1).

Step 3. Determination of the degrees of importance of ERs in relation toCRs, with linguistic terms

In this step, the degree of importance of the engineering requirements (ERs) inrelation to each group of CRs is determined, assuming initially that there is nodependence between the ERs. That is, for each CR, a matrix will be generated,relating its links to the ERs.

The calculation of the degrees of relative importance of ERs in relation to eachCR form the matrix W2, presented in a general way in Table 3.

Table 3: The W2 Matrix - Pairwise comparison of CRs in relation to ERs

W2 CR1 CR2 ... CRn

ER1 W11 W12 ... W1n

ER2 W21 W22 ... W2n

... ... .. ... ...

ERm Wm1 Wm2 ... Wmn

Page 8 of 20


Step 4. Construction of the CRs’ interdependence matrix, with linguisticterms

The W3 matrix of interdependence (or internal dependencies) between customerrequirements (CRs) is constructed in this step. To do this, it is necessary todetermine the interdependence between CRs in relation to each CR, indicatingthe degree of relative importance between them, according to the generic matrixshown in Table 4.

Table 4: The W3 Matrix - CRs’ interdependence matrix, with linguistic terms

CRi CR1 ... CRn

CR1 1 1 1 ... ... ... ... ... ...

... ... ... ... 1 1 1 ... ... ...

CRn ... ... ... ... ... ... 1 1 1

Step 5. Construction of the ERs’ interdependence matrix, with linguisticterms

As in the previous step, the construction of the interdependence matrix betweenthe ERs takes place in an analogous way and from the internal dependenciesbetween the ERs. The degrees of relative importance are determined for eachER. Finally, the matrix W4 of internal dependence between the ERs is filled withthe respective eigenvectors (Table 5).

Table 5: The W4 Matrix - ERs’ interdependence matrix, with linguistic terms

ERi ER1 ... ERn

ER1 1 1 1 ... ... ... ... ... ...

... ... ... ... 1 1 1 ... ... ...

ERn ... ... ... ... ... ... 1 1 1

Step 6. Internal prioritization of CRs

In this step, the interdependent priorities of customer requirements (RCs) are generated by calculating wC according to Eq. (1).

(1)

Step 7. Internal prioritization of ERs

In step 7, the interdependent priorities of the engineering requirements (ERs) are

generated by calculating WA according to Eq. (2):

(2)

Page 9 of 20


Step 8. Final prioritization of ERs in light of prioritized CRs

In step 8, the final prioritization of ERs ( wANP ) is obtained by Eq. (3):

. (3)

APPLICATION OF THE FUZZY-ANP-QFD MODEL: PROJECT OF A NEWTHERMAL VISION MONOCULAR

Aiming to demonstrate the applicability of the conceptual model presented in theprevious Section, the results of an empirical study developed within the contextof the COBRA 2020 Project are presented and discussed, focusing on the designof a new thermal vision monocular.

Choice of a new defense product for the purpose of the empirical study

The design of a new thermal vision monocular was selected to demonstrate the applicability of the conceptual model presented in Section 3, in function of the following conditions:

• It was just at the beginning of the project, and the R&D teams were verymotivated to test the conceptual model;

• Opportunity to provide objective specifications regarding the newmonocular design for potential suppliers, incorporating the perceptions of R& D teams and future users;

• Open up a R&D line at the Army Technological Center, with specialistsdedicated to the development of thermal vision monoculars aligned to theobjectives of the COBRA 2020 Project.

We present and discuss the results of the empirical study following the eightstages of the conceptual model presented in the previous Section.

Step 1. Definition of customer requirements (CRs) and association with engineering requirements (ERs) of the new thermal vision monocular

The definition of the customer requirements (CRs) of the new thermal visionmonocular was carried out by direct consultation with future users of the focusedproduct (combatants). The engineering requirements (ERs) were defined andgrouped by hearing members of the R&D teams involved in this development,based on a reference document published in the Army Bulletin, entitled "JointOperational Requirements - ROC 10/2012". It should be noted that the definitionof the Joint Operational Requirements was aligned with the Army Doctrine andOperational Guidelines applicable to all developments under the COBRA 2020Project (Brasil, 2014).

The first result of Step 1 was a list of 19 customer requirements (CRs), whichwere analyzed and grouped into four categories:

Page 10 of 20


• CR1 – Functionalities;• CR 2 – Ease to use;• CR 3 – Durability;• CR 4 – Possibility of couplings.

The second result was a list of 42 engineering requirements (ERs) groupedaccording to the five thermal vision monocular subsystems as follows:

• ER1 – Engineering requirements of the monocular body; • ER2 – Engineering requirements of the objective I; • ER3 – Engineering requirements of the processing and control system;• ER4 – Engineering requirements of the eyepiece; • ER5 – Engineering requirements of the LCD system.

The customer requirements (CRs) and engineering requirements (ERs) wereassociated as shown in Table 6.

Table 6: Customer requirements (CRs) associated with engineering requirements(ERs) of the new thermal vision monocular

Customer requirements (CRs)

Engineering requirements (ERs)ER1 - Monocular body

ER2 - Objective I

ER3 - Processing and control system

ER3 – Eyepiece

ER3 - LCD system

CR1 – Functionalities X X X X

CR2 - Ease to use X X X X

CR3 – Durability X X X

CR4 - Possibility of couplings

X

Step 2. Determination of the degrees of importance of CRs, with linguistic terms

In this step, a pairwise comparison matrix of the CRs was build based onjudgments of selected potential users (combatants) about the degree ofimportance of each CR, using linguistic terms, with fuzzy triangular numbers(Table 7).

Table 7: Pairwise comparison matrix of the CRs of the new thermal visionmonocular

CRsCR1 CR 2 CR 3 CR 4

l m u l m u l m u l m u

CR1 1 1 1 1 1/2 2 2 1/2 1/2 1 1 1/2 2 2 1/2 3

CR2 2/5 1/2 2/3 1 1 1 1/2 2/3 1 2 2 1/2 3

CR3 2/3 1 2 1 1,5 2 1 1 1 2 2 1/2 3

CR4 1/3 2/5 1/2 1/3 2/5 1/2 1/3 2/5 1/2 1 1 1

From the pairwise comparison matrix, we can obtain the eigenvector w1,calculated from the degree of importance of the customer requirements (CRs),through an expanded analysis of the fuzzy AHP method.

Page 11 of 20


DUE TO A large number of interactions, we chose to determine the relativeweights of each pAirwise comparison matrix, through an algorithm developed inthe Matlab platfoRM.

Step 3. Determination of the degrees of importance of ERs in relation toCRs, with linguistic terms

At this stage, it was assumed that there was no dependence between the ERs.The degrees of importance of the five groups of engineering requirements weredetermined in relation to each group of CRs.

Table 8 presents the pairwise comparison between ER2, ER3, ER4, AND ER5,which were the alternatives related to the criterion CR1 - ‘Functionalities'. Theweights’ column of relative importance contains the weights of relativeimportance, as explained in Section 3.

Table 8: Relative importance of ERs in relation to CR 1 - Functionalities

CR1

ER2 ER 3 ER 4 ER 5 Weights of relative importance

l m u l m u l m u l m u

ER2

1 1 1 2/5

1/2

2/3

1/2 1 1

1/2 2 21/2 3 0.2271

ER3

11/2 2 2

1/2 1 1 1 2 21/2 3 2 2

1/2 3 0.5284

ER4

2/3 1 2 1/3

2/5

1/2 1 1 1 2 2

1/2 3 0.2445

ER5

1/3 2/5 1/2 1/

3 2/5

1/2

1/3 2/5 1/2 1 1 1 0

Analogously, the degrees of importance concerning the other ERs concerning CR2, CR3 and CR4

were determined (Tables 9, 10 and 11).

Table 9: Relative importance of ERs in relation to CR 2 – Ease to use

CR2ER1 ER2 ER4 ER5 Weights of

relative importancel m u l m u l m u l m u

ER1 1 1 1 2/5 1/2 2/3 1/2 2/3 1 1/2 2/3 1 0.0684

ER2 1 1/2 2 2 1/2 1 1 1 1/2 1 1 1/2 2 2 1/2 3 0.4264

ER4 1 1 1/2 2 2/3 1 2 1 1 1 2 2 1/2 3 0.3945

ER5 1 1 1/2 2 1/3 2/5 1/2 1/3 2/5 1/2 1 1 1 0.1107

Page 12 of 20


Table 10: Relative importance of ERs in relation to CR 3 – Durability

CR3ER1 ER2 ER4 Weights of

relative importancel m u l m u l m u

ER1 1 1 1 2/5 1/2 2/3 1/2 2/3 1 0.1668

ER2 1 1/2 2 2 1/2 1 1 1 1/2 1 1 1/2 0.4405

ER4 1 1 1/2 2 2/3 1 2 1 1 1 0.3926

Table 11: Relative importance of ERs in relation to CR 5 – Possibility of couplings

RC4ER1 Weights of

relative importancel m u

ER1 1 1 1 1

The calculation of the degrees of relative importance of ERs in relation to eachCR forms the W2 Matrix, presented in Table 12.

Table 12: W2 Matrix – Pairwise comparison of CRs in relation to ERs of the newthermal vision monocular

W2 CR1 CR2 CR3 CR4

ER1 0 0.068 0.167 1

ER2 0.227 0.426 0.441 0

ER3 0.528 0 0 0

ER4 0.245 0.395 0.393 0

ER5 0 0.111 0 0

Step 4. Construction of the CRs’ interdependence matrix, with linguisticterms

From the analysis of interdependencies between the CRs, the degrees of relativeimportance were calculated for CR1 and presented in Table 13. For CR2, theresults are reported in Table 14 and for CR3 in Table 15.

Table 13: Relative importance of CRs in relation to CR1 - Functionalities

CR1CR1 CR3 CR4 Weights of relative

importancel m u l m u l m u

CR1 1 1 1 1/2 1 1 1/2 2 2 1/2 3 0.5

CR3 2/3 1 2 1 1 1 2 2 1/2 3 0.5

CR4 1/3 2/5 1/2 1/3 2/5 1/2 1 1 1 0

Page 13 of 20


Table 14: Relative importance of CRs in relation to CR2 – Easy to use

CR2CR2 CR3 CR4 Weights of relative

importancel m u l m u l m u

CR2 1 1 1 1/2 2/3 1 2 2 1/2 3 0.43

CR3 1 1,5 2 1 1 1 2 2 1/2 3 0.57

CR4 1/3 2/5 1/2 1/3 2/5 1/2 1 1 1 0

Table 15: Relative importance of CRs in relation to CR3 – Durability

CR3CR2 CR3 Weights of relative

importancel m u l m u

CR2 1 1 1 1/2 2/3 1 0.33

CR3 1 1,5 2 1 1 1 0.67

Finally, the resulting W3 Matrix is presented in Table 16. The column referring toCR4 – Possibility of couplings is null because this criterion is not subordinated tothe other CRs.

Table 16: W3 Matrix – Interdependencies between CRs

W3 CR1 CR2 CR3 CR4

CR1 1/2 0 0 0

CR2 0 3/7 1/3 0

CR3 1/2 4/7 2/3 0

CR4 0 0 0 0

Step 5. Construction of the ERs’ interdependence matrix, with linguisticterms

As in the previous Step, the construction of the interdependence matrix betweenthe ERs took place in an analogous way and from the analysis of internaldependencies between the ERs. The degrees of relative importance weredetermined for each ER (Tables 17, 18, 19, and 20). Finally, the W4 Matrix wasfilled with the respective eigenvectors (Table 21).

Table 17: Relative importance of ERs in relation to ER1 - Engineeringrequirements of the monocular body

ER1ER1 ER2 ER3 ER4 ER5

l m u l m u l m u l m u l m u

ER1 1 1 1 2/5 1/2 2/3 2/5 1/2 2/3 1/2 1/21

1/21/2 1

11/2

ER21

1/22 2 1/2 1 1 1 2/5 1/2 2/3 1/2 1

11/2

2 2 1/2 3

ER31

1/22 2 1/2 1 1/2 2 2 1/2 1 1 1 2 2 1/2 3 2 2 1/2 3

Page 14 of 20


ER4 1 1 1/2 2 2/3 1 2 1/3 2/5 1/2 1 1 1 2 2 1/2 3

ER5 1 1 1/2 2 1/3 2/5 1/2 1/3 2/5 1/2 1/3 2/5 1/2 1 1 1

Page 15 of 20


Table 18: Relative importance of ERs in relation to ER3 - Engineeringrequirements of the processing and control system

ER3ER2 ER3

l m u l m u

ER2 1 1 1 2/5 1/2 2/3

ER3 1 1/2 2 2 1/2 1 1 1

Table 19: Relative importance of ERs in relation to ER4 - Engineeringrequirements of the eyepiece

ER4ER4 ER5

l m u l m u

ER4 1 1 1 2 2 1/2 3

ER5 1/3 2/5 1/2 1 1 1

Table 20: Relative importance of ERs in relation to ER5 - Engineeringrequirements of the LCD system

EP5ER3 ER5

l m u l m u

ER3 1 1 1 2 2 1/2 3

ER5 1/3 2/5 1/2 1 1 1

Finally, the W4 Matrix is filled with the ERs’ eigenvectors (Table 21).

Table 21: W4 Matrix - ERs’ interdependence matrix of the new thermal vision monocular

W4 ER1 ER2 ER3 ER4 ER5

ER1 0.030 0 0 0 0

ER2 0.273 0 0 0 0

ER3 0.448 0 1 0 1

ER4 0.248 0 0 1 0

ER5 0 0 0 0 0

Step 6. Internal prioritization of CRs

In this step, the interdependent priorities of customer requirements (CRs) were generated by calculating wC according to Eq. (1).

Page 16 of 20


Step 7. Internal prioritization of ERs

In step 7, the interdependent priorities of the engineering requirements (ERs)

were generated by calculating WA according to Eq. (2):

Step 8. Final prioritization of ERs in light of prioritized CRs

In step 8, the final prioritization of ERs ( wANP ) was obtained by Eq. 3:

The wANP vector indicates the final prioritization of engineering requirements

(ERs) in the light of prioritized customer requirements (CRs), as shown in Table 22.

Table 22: Final prioritization of ERs in the light of prioritized CRs of the newthermal vision monocular

Engineering requirementsRelative

importance Final prioritization

ER1 – Engineering requirements of the monocular body

0.0034 4th

ER2 – Engineering requirements of the objective I

0.0307 2nd

ER3 – Engineering requirements of the processing and control system

0.1785 3rd

ER4 – Engineering requirements of the eyepiece

0.3919 1st

ER5 – Engineering requirements of the LCD system

0 5th

Discussion of empirical results

The final prioritization of engineering requirements (ERs) in the light of customer

requirements (CRs) of the new thermal vision monocular indicated that: (i)

eyepiece requirements (ER4) came first; (ii) the engineering requirements of the

objective I (basic mode) came in the second position; (iii) the engineering

Page 17 of 20


requirements of the processing and control system (ER3) came in third place; (iv)

the engineering requirements of the monocular body (ER1), in fourth; and (v)

finally, the LCD system requirements (ER5) in the last position.

The innovative aspects of the fuzzy-ANP-QFD approach proposed for new

defense product design are highlighted, namely:

• The adoption of the proposed approach will contribute to guaranteeing thefulfillment of the duality requirements conferred to some new defenseproducts from its conception;

• As a support tool in prioritization processes, it may subsidize decisionmaking in situations where it is necessary to define between thedevelopment or acquisition of subsystems of a new defense product;

• It may help managers in prospecting emerging technologies for newdefense product design when integrated to Design Thinking tools.

CONCLUDING REMARKS

From the empirical results presented in this work, we conclude that theconceptual model based on the fuzzy-ANP-QFD approach allows a more realisticprioritization for the engineering requirements (ERs) in the light of the customerrequirements (CRs), than with the fuzzy-AHP. Although the proposed model hassimplified the number of ERs and CRs, by grouping them in a smaller number ofentries, it is important to emphasize that the model does not impose limitationsin this sense.

Taking into account that new defense product design must be aligned to ArmyDoctrine and Operational Guidelines, which emphasize operational conceptsinstead of customer needs, the future adoption of the proposed model can bringan innovative perspective and potential benefits for this relevant field.

REFERENCES

Abdolshah, M., and Moradi, M., (2013), Fuzzy quality function deployment: ananalytical literature review. Journal of Industrial Engineering, 2013, 1-11.

Akao, Y. (Ed), (1990), Quality function deployment: integrating customerrequirements into product design. Cambridge: Productivity Press.

Akao, Y., and Mazur, G. H., (2003), The leading edge in QFD: past, present, andfuture. International Journal of Quality & Reliability Management, 20 (1), 21-35.

Brasil. Ministério da Defesa. Exército Brasileiro, (2014), Projeto CombatenteBrasileiro 2020. Escopo. Brasília: MD.

Chen, S-H., Chen, F-Y., and Yeh, T-M., (2015), Integration quality functiondeployment and analytic hierarchy process methods to improve service quality.American Journal of Food Technology, 10 (2), 85-92.

Clausing, D., (1994), Total quality development. New York: ASME Press, 506 p.

Page 18 of 20


Dai, J., and Blackhurst, J., (2012), A four-phase AHP-QFD approach for supplierassessment: A sustainability perspective. Int. J. Prod. Res., 50, 5474-5490.

Emrouznejad, A., and Ho, W., (2017), Fuzzy analytic hierarchy process. BocaRaton, FL: CRC Press LLC.

Güngor, Z., Delice, S.E., and Kesen, E.K., (2011), New product design using FDMSand FANP under fuzzy environment. Applied Soft Computing, 11 (11), 3347-3356.

Hauser, J.R., Clausing, D., (1998), House of quality. Harvard Business Review, 66,63-67.

Ho, W., He, T., Lee, C., and Emrouznejad, A., (2012), Strategic logisticsoutsourcing: an integrated QFD and fuzzy AHP approach, Expert Systems withApplications 39, 10841–10850.

Kahraman, C., Ertay, T., and Büyüközkan, G., (2006), A fuzzy optimization modelfor QFD planning process using analytic network approach. European Journal ofOperational Research, 171, 390-411.

Kam, K., and Yuen, F. (2014), A hybrid fuzzy quality function deploymentframework using cognitive network process and aggregative grading clustering:An application to cloud software product development. Neurocomputing, 142, 95-106.

King, R., (1987), Listening to the voice of the customer: using the quality functiondeployment system. National Productivity Review, 6 (3), 277-281.

Lee, Y.C., Sheu, L.C., and Tsou, Y.G., (2008), Quality function deploymentimplementation based on fuzzy Kano model: an application in PLM system.Computers and Industrial Engineering, 55, 48–63.

Liu, H-T. , (2009), The extension of fuzzy QFD: from product design to partdeployment. Expert Systems with Applications, 36, 11131–11144.

Liu, H-T., (2011), Product design and selection using fuzzy QFD and fuzzy MCDMapproaches. Applied Mathematical Modelling, 35 (1), 482-496.

Maritan, D., (2015), Practical manual of quality function deployment. Genebra:Springer International Publishing Switzerland.

Mayyas, A., Shen,Q., Mayyas, A., Abdelhamid, M., Shan, D., Qattawi, A., andOmaret, M., (2011), Using quality function deployment and analytical hierarchyprocess for material selection of body-in-white. Materials & Design, 32 (5), 2771-2782.

Raharjo, H., Brombacher, A. C., And Xie, M., (2008), Dealing with subjectivity inearly product design phase: A systematic approach to exploit Quality FunctionDeployment potentials. Computers & Industrial Engineering, 55, 253-278.

Saaty, T.L., (1996), The analytic network process: decision making withdependence and feedback. Pittsburgh: RWS Publications.

Saaty, T.L., (1980), The analytic hierarchy process. New York: McGraw-Hill.

Page 19 of 20


Tu, C.S., Chang, C.T., Chen, K.K., and Lu, H.A., (2010), Applying an AHP-QFDconceptual model and zero-one goal programming to requirement-based siteselection for an airport cargo logistics center. International Journal of InformationManagement, 21, 407-430.

Vanegas, L.V., and Labib, A.W., (2001), A fuzzy quality function deployment(FQFD) model for deriving optimum targets. International Journal of ProductionResearch, 39 (1), 99–120.

Vinodh, S., and Chintha, S., (2011), Application of fuzzy QFD for enablingleanness in a manufacturing organization. International Journal of ProductionResearch, 49 (6), 1627–1644.

Wang, H., Xie, M., and Goh, T.N., (1998), A comparative study of the prioritizationmatrix method and the analytic hierarchy process technique in quality functiondeployment. Total Quality Management, 9 (6), 421-430.

Wang, C.H., (2010), An integrated fuzzy multi-criteria decision-making approachfor realizing the practice of quality function deployment, in Proceedings of theIEEE International Conference on Industrial Engineering and EngineeringManagement (IEEM ’10), pp. 13–17, December 2010.

Weng, M.C., Hsiao, J.M., and Tsai, C.H., (2009), Fuzzy analytical approach toprioritizing design requirements in quality function deployment. Journal ofQuality, 16 (1), 16, 61-71.

Younesi, M., and Roghanian, E., (2015), A framework for sustainable productdesign: a hybrid fuzzy approach based on Quality Function Deployment forEnvironment. Journal of Cleaner Production, 108 (part A), 385-394.

Zadeh, L. A., (1965), Fuzzy sets. Information and Control, 8, 177-200.

Zaim, S., Sevkli, M., Camgöz-Akdağ, H., Demirel, O.E., Yayla. A.Y., and Delen, D.,(2014), Use of ANP weighted crisp and fuzzy QFD for product development.Expert Systems with Applications, 41, 4464–4474.

Zhou, M., (1998), Fuzzy logic and optimization models for implementing QFD.Computers and Industrial Engineering, 35 (1–2), 237–240.

Page 20 of 20

FUZZY QUALITY FUNCTION DEPLOYMENT (FUZZY-QFD) …

Documents