برخورد با ذهنیت در مرحله طراحی محصول اولیه: رویکرد سیستماتیک به بهره برداری از پتانسیل های گسترش کارکرد کیفیت

Quality Function Deployment (QFD), as a customer-driven tool, is generally used in the early phase of new or improved products/services design process, and therefore most of the input parameters are highly subjective in nature. The five major input components of the QFD, which are laid in the House of Quality (HOQ), namely, the customer requirement, the technical attribute, the relationship matrix, the correlation matrix, and the benchmarking information, play a central role in determining the success of QFD team. Accurate numerical judgment representations are of high importance for the QFD team to fill in the values of each of those components. In this paper, a generic network model, based on Analytic Network Process (ANP) framework, will be proposed to systematically take into account the interrelationship between and within those components simultaneously and finally derive their relative contribution. In particular, with respect to a rapidly changing market, the incorporation of the new product development risk, the competitors’ benchmarking information, and the feedback information into the network model may be considered as a novel contribution in QFD literature. Not only does this network model improve the QFD results’ accuracy, but it also serves as a generalized model of the use of ANP in QFD with respect to the previous research. A simple illustrative example of the proposed network model will be provided to give some practical insights.

Superior product design, potential for breakthrough innovation, low project and product cost, shorter lead time, better communication of cross-functional teamwork, and increased customer satisfaction and market share are among many other advantages that make Quality Function Deployment (QFD) an important, and yet unique, tool for successful new product development (Chan and Wu, 2002a, Griffin and Hauser, 1992, Hauser and Clausing, 1988, Presley et al., 2000 and Xie et al., 2003). It enables firms in making strategic decisions while ensuring full knowledge of the customer, the technology, and with the team’s support (Hauser, 1993). Essentially, the QFD starts and ends with the customer. The Voice of Customer (VOC) (Griffin & Hauser, 1993) is the main driver and will be propagated through all subsequent downstream processes, and as a result, greater customer satisfaction is created in the end product/service. According to a study by Griffin (1992), the two most critical factors that determine the QFD’s successful use in providing definite strategic product development benefits are the high commitment of all team members in all functional areas, and the paradigm that treats QFD as a cross-functional investment in people and information. Since the focus of the QFD is on the early phase of new products/services design or redesign process, most of the input parameters are therefore highly subjective in nature (Kim et al., 2007 and Xie et al., 2003). Based on the survey results over 400 companies in the US and Japan, Cristiano, Liker, and White (2000) showed that the QFD analysis may only require a simple and practical decision aid based upon the experience and judgment of the team. This is mainly attributed to the fact that the QFD was born out of an industry need for ensuring design quality. Hence, the accuracy level of these subjective experience and judgment will significantly determine the quality of the QFD results. In view of this, a method or approach that is capable to systematically analyze and accurately quantify those subjective experience and judgments of the QFD team is highly required. In the literature, the Analytic Hierarchy Process (Saaty, 1983 and Saaty, 1994), of which generalized form is called the Analytic Network Process (Saaty, 1996), is known as one of the most powerful management science tools to serve this purpose. The AHP/ANP has been widely accepted as a realistic, flexible, simple, and yet mathematically rigorous modeling technique in multiple criteria decision making field (Liberatore, 1987, Saaty, 1986, Sarkis and Sundarraj, 2006 and Vaidya and Kumar, 2006). The AHP/ANP framework can be considered as a powerful and necessary tool for making any strategic decision since it is capable of taking into consideration multiple dimensions of information from multi-party, either qualitative or quantitative, into the analysis (Dyer and Forman, 1992, Meade and Presley, 2002 and Meade and Sarkis, 1998). In using the AHP in new product development field, Calantone, Di Benedetto, and Schmidt (1999) wrote that “the AHP helps managers make more rational decisions by structuring the decision as they see it and then fully considering all of the information”. In other words, the AHP/ANP effectively facilitates managers in quantifying their subjective judgments, experience, and knowledge of the complex system in an intuitive and natural way ( Dey, 2004 and Mustafa and Al-Bahar, 1991) by systematically taking into account all the relevant factors and their relative effects as well as interactions simultaneously. The trend of the AHP/ANP’s use to assist QFD practitioners develop new or redesign existing products has been remarkably increasing (Armacost et al., 1994, Cohen, 1995, Lu et al., 1994 and Zakarian and Kusiak, 1999). Recently, Ho (2007) gave a review of the use of other methods in combination with the AHP, and found that the combination of the AHP and QFD is one of the most commonly used technique in the literature. Furthermore, due to its very exceptional strength in addressing the inner-relationship and interrelationship among the QFD components, the ANP has also increasingly been used in QFD recently (see Büyüközkan et al., 2004, Ertay et al., 2005, Kahraman et al., 2006, Karsak et al., 2002, Pal et al., 2007, Partovi, 2006 and Partovi, 2007). The use of ANP in QFD, in general, can be categorized into two types. The first type, of which model has been used by quite many researchers (Büyüközkan et al., 2004, Ertay et al., 2005, Kahraman et al., 2006, Karsak et al., 2002 and Pal et al., 2007), is mainly based on the network model described in Saaty and Takizawa (1986). Compared to the recent development of the ANP method, it might be considered as rather preliminary (see Section 2.2). While the second type, which can be considered as a better advancement of the use of ANP in QFD, employs the network model proposed recently by Partovi (Partovi, 2006 and Partovi, 2007). However, the model is still rather restricted in the sense that it uses ANP in addressing only two elements of House of Quality (HOQ), namely, the relationship matrix and the correlation matrix (the roof of HOQ). To fill in the niche of using ANP in QFD more effectively, this paper therefore proposes a generic network model which serves as a generalized model from the previous research work. In particular, it takes into account the product design risk, competitors’ benchmarking information, and feedback information among the factors involved. It is hoped that by using the proposed network model, the accuracy of the QFD results can be further enhanced, particularly with respect to a constantly changing market. In other words, by providing an effective way of quantifying and analyzing QFD team’s subjective experience, knowledge, and judgments systematically, this paper enables QFD practitioners to exploit more potentials of QFD as a powerful early product/service design tool. In the next section (Section 2), a brief review of the AHP/ANP and its use in QFD will be provided. Section 3 describes the significance of some factors that are selected to be included in the proposed network model. Then, the proposed network model, which is the key contribution of this paper, is elaborated in Section 4. To give some practical insights when using the proposed network model, an illustrative example was developed (Section 5), as a result from an intensive interview with people who are knowledgeable in the design process of consumer electronic products. Finally, a discussion of the proposed method as well as possible future work will be described in Section 6.

The purpose of this paper, with respect to a constantly changing market, was to better deal with subjectivity in early product design phase by exploiting QFD potentials using the ANP approach. The main contribution of this paper lies in the proposed generic ANP-based network model which can be used to assist QFD practitioners in accurately quantify their subjective judgments and experience in a systematic fashion. Interestingly, not only can the network model address all elements in the HOQ, which may therefore serve as a substitutive procedure to the traditional QFD method, but it also takes into account other important factors in the product design context, such as the new product development risk. In general, the proposed network model has formally taken into account most of the crucial factors in new product design simultaneously. The relative contributions of each factor and their inner-action as well as interaction have been systematically incorporated. In particular, with respect to the existing QFD literature, the inclusion of the new product development risk, the competitors’ benchmarking information, and the feedback information into the network model, which is of considerable importance to new product development’s success, can be regarded as the novel contribution of this study. Thus, not only does this network model improve the QFD results’ accuracy, it also serves as a generalized model of the use of ANP in QFD from the previous research (Büyüközkan et al., 2004, Ertay et al., 2005, Kahraman et al., 2006, Karsak et al., 2002, Partovi, 2006, Partovi, 2007 and Pal et al., 2007). Some advantages of using the proposed network model may include the reduction of human judgment error, transparent evaluation, and improved efficiency. More importantly, the flexibility of the QFD in adapting to the constantly changing environment can be significantly improved as a sensitivity analysis to dynamically evaluate the network model can be carried out at any time. As with other ANP applications, a major possible drawback is the trade-off between the model complexity and the required time to complete the pairwise comparisons (Meade and Presley, 2002, Ravi et al., 2005, Sarkis and Sundarraj, 2006 and Shang et al., 2004). Nevertheless, when a substantial amount of risk, including financial risk, is involved, then a systematic and structured analysis of dealing with the problem can be fully justified. Note that the numerical outcomes of the method are less important than the systematic thinking environment it offers. With regard to the implementation and to avoid a too mechanistic application of the proposed network model, there are some points worth noting, as one of the authors learnt from the interview process with the design experts. First, the terms used in the questionnaire for each cluster in the model should be clearly explained, for example, what it means by ‘intuitiveness’ should be clearly defined beforehand (see Appendix A). Second, the meaning of Saaty’s fundamental scale used in the pairwise comparison for eliciting decision maker’s judgments should also be explained clearly (Appendix A). These two points are the most relevant operational difficulty that QFD team might encounter when using the model. In addition, the proposed network model can only be used for representing one House of Quality (HOQ), for example, in the illustrative example used in this paper, it is for representing the first HOQ. How this network model can be used for representing other HOQ, that is, the second, the third, and so forth remains a challenging topic for subsequent work. The proposed network model, to a certain extent, is also limited since it has not taken into account all possible factors. However, this also, at the same time, shows the versatility of the model, which allows further expansion to suit the condition of a particular company. In general, possible avenues for future work can be channeled in two directions, namely, the practical aspect and the theoretical aspect. From a practical standpoint, the incorporation of more real-world case studies to showcase the effectiveness of the proposed model would be a significant contribution. While from a theoretical standpoint, a comparison study of the proposed model’s effectiveness with other QFD methodologies remains an interesting area for future work.