درجه بندی اهمیت فنی در گسترش کارکرد کیفیت فازی با ادغام عادی سازی فازی و متوسط ​​وزنی فازی

Fuzzy quality function deployment (QFD) has been extensively used for translating customer requirements (CRs) into product design requirements (DRs) in fuzzy environments. Existing approaches, however, for rating technical importance of DRs in fuzzy environments are found problematic, either incorrect or inappropriate. This paper investigates how the technical importance of DRs can be correctly rated in fuzzy environments. A pair of nonlinear programming models and two equivalent pairs of linear programming models are developed, respectively, to rate the technical importance of DRs. The developed models are examined and illustrated with two numerical examples.

Quality function deployment (QFD) [1] is a methodology for translating customer requirements (CRs), i.e. the voice of the customer (VoC), into product design requirements (DRs). In this translating process, a large number of subjective judgments have to be made by both customers and QFD team members. Due to uncertainty and vagueness involved in subjective judgments, fuzzy logic has been widely suggested for better capturing the relative importance of CRs and the relationships between CRs and DRs as well as the correlations among DRs. As a result, fuzzy QFD has been developed, researched and extensively applied [2], [3], [4], [5], [6], [7], [8], [9], [10] and [11]. For fuzzy QFD, one of the key issues is to derive the technical importance ratings of DRs in fuzzy environments and prioritize them so that limited resources such as budget can be reasonably or optimally allocated within DRs in terms of their priorities. Existing approaches for rating the technical importance of DRs in fuzzy environments are found problematic, either incorrect or inappropriate. Therefore, there is a need to develop a correct methodology for rating the technical importance of DRs. This paper investigates how the technical importance of DRs can be correctly rated in fuzzy environments. A pair of nonlinear programming (NLP) models is developed to correctly rate the technical importance of DRs in fuzzy environments, which is then broken down into two equivalent pairs of linear programming (LP) models for solution. The paper is organized as follows. Section 2 gives a brief introduction to fuzzy sets and fuzzy weighted average that are or will be used in fuzzy QFD. Section 3 presents a literature review on the formulas and approaches for rating the technical importance of DRs in fuzzy environments and points out their incorrectness or inappropriateness. Section 4 develops correct NLP models for rating the technical importance of DRs. Section 5 shows how the NLP models can be simplified as two equivalent pairs of LP models for solution. The developed models, linear and nonlinear, are examined and illustrated with two numerical examples in Section 6. The paper concludes in Section 7.

As a customer-oriented methodology, QFD has been widely applied to improve product quality to achieve higher or maximum customer satisfaction. To achieve maximum customer satisfaction, limited organizational resources such as budget have usually to be optimally allocated among DRs in terms of their technical importance ratings. Without correct technical importance ratings of DRs, it would be impossible to optimally allocate limited organizational resources among them to achieve maximum customer satisfaction. This requires the development of a method that can correctly rate the technical importance of DRs. In this paper we have investigated how to correctly rate the technical importance of DRs in fuzzy QFD using αα-level sets. We have presented a literature review on existing approaches for rating technical importance of DRs in fuzzy environments and pointed out their incorrectness or inappropriateness. We have then developed a pair of NLP models and two pairs of LP models, respectively, for rating the technical importance of DRs in fuzzy environments. We have proved the equivalence of the two pairs of LP models and the pair of NLP models. The developed models have finally been examined and illustrated with two numerical examples. The numerical examinations have clearly revealed that the developed models can correctly and accurately rate the technical importance of DRs in fuzzy environments. In comparison with existing approaches for rating the technical importance of DRs in fuzzy environments, the developed models have some significant merits such as being able to rate the technical importance of DRs in fuzzy environments accurately through αα-level sets, producing normalized fuzzy technical importance ratings for DRs, offering two alternative options, linear and nonlinear programming, for technical importance rating. It is believed that the developed models make a good contribution to fuzzy QFD and lay a solid theoretical foundation for its development and applications. The future research work is to combine the developed models and limited resources to set targets for DRs. A belief-rule based (BRB) methodology will be developed.