0 G1 1 0
1
Probability and Possibility
Fuzzy requirement
Target achievement
Figure 6.4: Benchmark achievement: fuzzy min G1
contrast to Wang and Chen’s [31, 136] work, our approach can catch the slight changes of DM’s requirement as well as coincide with the intuition of DM.
Multi-criteria decision analysis (MCDA) problems can be categorized into two classes:
discrete and continuous MCDA [127], in this study we focused on discrete MCDA (MADA).
In the context of continuous MCDA (multi-objective decision analysis), Ogryczak and Sliwi´´ aski [107] proposed using the OWA operator to solve linear programming problems.
Furthermore, it is practical to consider that there are different importances and priorities of objectives [28,60,86,87]. In this regard, we believe that our proposed prioritized OWA operator will provide a general and convenient tool for multi-objective decision making with multiple priorities. However, this is left for the future work.
Chapter 7
Kansei Evaluation Based on
Prioritized Multi-Attribute Fuzzy Target-Oriented Decision Analysis
Abstract: In this chapter, we focus on the evaluation problems using Kansei data, tak-ing consumers’ preferences on Kansei attributes into consideration. This chapter aims at proposing and developing a Kansei evaluation model based on multi-attribute target-oriented decision analysis. To do so, firstly, like the traditional Kansei evaluation method, a preparatory experiment study is conducted in advance to select Kansei attributes by means of semantic differential (SD) method. In order to obtain Kansei data of products, a number of subjects are selected to assess products regarding these Kansei attributes.
Differed from the previous research, linguistic variables are used to represent the uncertain assessments. Secondly, these Kansei data are used to generate Kansei profiles for eval-uated products by means of the voting statistics. Thirdly, as consumers’ preferences on Kansei attributes vary from people to people and the consumers may have prioritization of Kansei attributes, we can view the current Kansei evaluation problem as a prioritized multi-attribute target-oriented decision analysis problem. Based on the proposed Kansei evaluation model, the consumers can select or choose the products according to their preferences.
7.1 Introduction
In today’s increasingly competitive market place, satisfying consumers’ needs and tastes has become a great concern of almost every company [59,68,129]. Consumers have put more emphasis not only on functional requirements of products, defined objectively, but also on psychological needs and feelings, by essence subjective [112]. Moreover, with the development of global markets and modern technologies, it is likely that many similar products will be functionally equivalent [68], thus consumers may find that it is difficult to distinguish and choose their desired product(s). In this regard, consumers’ psychological needs and feelings must be considered in choice of products.
Kansei engineering has been developed as a methodology to deal with consumers’
subjective impressions (called Kansei in Japanese) regarding a product into the design elements of a product [103, 104, 105]. There is no corresponding term to Kansei in English. The term Kansei is imbedded in the Japanese culture in a way that is difficult to translate into words. A specific Kansei arises when a human is subjected to an artifact in a certain environmental context [118]. Kansei may be easier to experience than define by a western person. Looking at picture or artifact may evoke a certain “good feeling”
that is difficult to describe. This is what Kansei is about. According to Nagamachi [103]
and Sch¨utte [118],
Kansei is an individual subjective impression from a certain artifact, envi-ronment or situation using all the senses ofsight, hearing, feeling, smell, taste, recognition and balance.
For building a Kansei database on psychological feelings regarding products, the most commonly used method is to choose Kansei words (bipolar subjective words) first, and then ask people to express their feelings using those words often by means of the semantic differential (SD) method [109] or its modifications [53,118].
In this chapter, we focus on the evaluation problems using Kansei data, taking con-sumers’ preferences on Kansei attributes into consideration. The evaluation would be of great help for marketing or recommendation purposes, and particularly in the era of e-commerce, where recommendation systems have become an important research area [4].
It should be emphasized that many studies of Kansei engineering or other consumer-oriented design techniques have involved an evaluation process, in which a design could be selected for production, e.g., [112].
This chapter aims at proposing and developing a Kansei evaluation model based on multi-attribute target-oriented decision analysis. To do so, firstly, like the traditional Kansei evaluation method, a preparatory experiment study is conducted in advance to select Kansei attributes by means of semantic differential (SD) [109] method. In order to obtain Kansei data of products, a number of subjects are selected to assess products regarding these Kansei attributes. Differed from the previous research, linguistic variables are used to represent the uncertain assessments. Secondly, these Kansei data are used to generate Kansei profiles for evaluated products by means of the voting statistics. Thirdly, as consumers’ preferences on Kansei attributes vary from people to people and the con-sumers may have prioritization of Kansei attributes, we can view the current Kansei evaluation problem as a prioritized multi-attribute target-oriented decision analysis prob-lem. In particular, three main types of fuzzy targets are defined to represent consumers’
uncertain preferences on Kansei attributes. Based on the fuzzy target-oriented decision
analysis model discussed in Chapter 4, we can obtain the satisfaction degrees (probabili-ties of meeting targets) regarding the Kansei attributes selected by consumers for all the evaluated products. And then, considering prioritization of the Kansei attributes, the so-called prioritized OWA aggregation operator discussed in Chapter 6 is used to aggre-gate the partial satisfaction degrees for the evaluated products. Based on the proposed Kansei evaluation model, the consumers can select or choose the products according to their preferences.
The rest of this chapter is organized as follows. In Section 7.2 we recall some basic knowledge of Kansei evaluation and give the motivations of our work. In Section 7.3 we put forward a Kansei evaluation model based the fuzzy target-oriented decision model discussed in Chapter 4 and the prioritized OWA aggregation operator proposed in Chapter 6. In Section7.4we give some discussions of the our model and give some effective analysis.
Finally, some concluding remarks are given in Section 7.5.