Fuzzy Extension for Kano's Model Using Bacterial Evolutionary Algorithm P. Fo1desi '- L.T. Koczy 2,3 J Botzheim 3 ' Department of Logistics and Forwarding, Szechenyi Istvan University H-9026, Gy6r, Egyetem ter 1. Hungary 2Department of Telecommunication, Szechenyi Istvan University H-9026, Gy6r, Egyetem ter 1. Hungary 3 Department of Telecommunications and Media Informatics, Budapest University of Technology and Economics, H- 11 17 Budapest, Magyar tudosok korfitja 2, Hungary Abstract particular satisfaction functions are partly linear and partly For designing and developing products/services it is vital to know non-linear and in a given market segment the overall the relevancy of the performance generated by each technical satisfaction of all customers is the optimum criteria, so within attribute and how they can increase customer satisfaction. budget limits set of technical attributes are to be determined Improving the parameters of technical attributes requires that can maximize the overall satisfaction. financial resources, and the budgets are generally limited. Thus the optimum target is to achieve maximum customer satisfaction within given financial limits. Kano's quality model classifies the II. KANO'S QUALITYMoDEL relationships between customer satisfaction and attribute-level performance and indicates that some of the attributes have a non- In his model [6] Kano distinguishes three types of product linear relationship to satisfaction, rather power-function should requirements which influence customer satisfaction in different be used. For the customers' subjective evaluation these ways when met (see Figure 1). relationships are not deterministic and are uncertain. This paper * Degressive or "must-be" requirements are basic proposes a method for fuzzy extension of Kano's model and criteria of a product. From a given point improving presents numerical examples that can prove the efficiency of the technical attributes by unit results in minor bacterial evolutionary algorithm in as well. Key words: quality, Kano's model, fuzzy, bacterial algorithm increment in satisfaction, on the other hand not fulfilling the requirements induces dissatisfaction I. INTRODUCTION ("negative satisfaction"). See function D in Fig. 1. * One-dimensional requirements. Customer satisfaction In the designing process of products/services, their technical is proportional to the level of fulfillment, the higher attributes must be determined so that the maximum customer level of fulfillment, the higher the customer satisfaction can be achieved within acceptable and reasonable satisfaction, and vice versa. See function L in Fig. 1. financial limits. Technical attributes have different effects on * Progressive or attractive/excitement factors: fulfilling the satisfaction. Kano explored [5,6] that the features and these requirement leads to more than proportional characteristics of these relationships differ from the point of satisfaction. See function P in Fig. 1. view of customers the utility functions are different as well. On the other hand customers requirements are not homogenous, p they are changing in time and also differences can be detected Customer even in the same market segment. Because of these differences satisfaction in the mathematical model for Kano's quality assessment it is worth applying fuzzy numbers instead of crisp values. Results have been devoted to the relationship between technical attributes and customer requirements in correlation terms [3, 4], or represented the uncertainty of budgeting by fuzzy measures [11]. Assuming linear relationship [9] and [1] analyzed the issue. Reference [8] considered it as a linear problem by introducing the customer satisfaction coefficient.// Reference [7] explored the asymmetric feature of the// relationship between attribute-level performance and overall// customer satisfaction and indicated indirectly that linear/ functions are not appropriate in each case. Application of fuzzy _________________________ logic for ranking technical attributes is presented in [12]. tcnclatiue that meet needs In this paper the proposed fuzzy extension for Kano' s model Fig.1. Kano's model of customer satisfaction handles the customers assessment as a fuzzy number, the 1-4244-11 58-0/07/$25.OO © 2007 IEEE. 147