An integrated method of rough set, Kano’s model and AHP for rating customer requirements’ final importance

https://doi.org/10.1016/j.eswa.2008.08.036Get rights and content

Abstract

Owing to the typical vagueness or imprecision of customer requirements (CRs) in product planning house of quality (PPHOQ), determining the final importance of CRs is very difficult. Combining rough set theory, Kano’s model, analytical hierarchy process (AHP), and scale method, an integrated method is proposed to obtain the final importance of CRs in PPHOQ. Firstly, by using relative reduction and relative core in rough set theory, a decision system is built to acquire CRs in PPHOQ. Secondly, based on relative positive field in rough set, the decision system is simplified and its corresponding new decision system is established to determine the fundamental importance ratings of CRs. Thirdly, by integrating scale method into AHP approach, calculating formulas of the importance rating of achieving the improvement ratio of satisfaction estimation of a CR are developed. Next, for every CR, based on a combination of its fundamental importance rating, the importance rating of achieving the improvement ratio of its satisfaction estimation, and “its sales point”, its final importance rating is determined. Finally, a case study is provided to illustrate the effectiveness of the presented method.

Introduction

Quality function deployment (QFD) is a widely used customer-driven design and manufacturing approach developed in Japan in the late 1960s. It has gained extensive international support (Chen et al., 2004, Chen et al., 2006, Bai and Kwong, 2003, Chan and Wu, 2002, Hauser and Clausing, 1988, Kim et al., 2000, Xie et al., 2003). Generally, it utilizes four sets of matrices called house of quality (HOQ) to translate CRs into engineering characteristics (ECs), subsequently, into parts characteristics, process plans, and production requirements (Fung et al., 2006, Luo et al., 2007, Tang et al., 2002, Tang et al., 2005). The first set of matrices called product planning house of quality (PPHOQ), is of fundamental and strategic importance in QFD system (Fung et al., 2002, Fung et al., 2003, Karsak, 2004, Park and Kim, 1998, Raharjo et al., 2006, Wasserman, 1993), since CRs for the product are identified, and incorporated the company’s competitive priorities, converted into ECs, and then the target levels for ECs of improved products are determined. In the complex decision process, determining the final importance rating of CRs is a crucial step (Armacost et al., 1994, Chan et al., 1999, Chan and Wu, 2005, Kim et al., 2007, Kwong and Bai, 2003, Wang, 1999). Based on these ratings, a company can purposely design and develop to match or exceed customer satisfaction of all competitors in target market and thus achieve more competitive advantages.

Generally, the determination of the final importance ratings of CRs in PPHOQ consists of four steps (Chan et al., 1999, Chan and Wu, 2005). These proposed hierarchical steps are described as follows:

  • Step 1.

    Identify customers and acquire their requirements.

  • Step 2.

    Determine the fundamental importance ratings of CRs.

  • Step 3.

    Identify competitors and conduct relative analysis.

  • Step 4.

    Determine the final importance ratings of CRs.

Since CRs are the driving force in the PPHOQ, considerable effects must be committed to properly capture those requirements to keep a company successful. According to Chan and Wu (2002), there are many methods available to collect CRs, including focus group, individual interviews, listening and watching, complaints, natural field contact, warranty data, feedback, affinity diagram, and cluster analysis. However, all of the approaches can not provide any quantitative method for acquiring CRs in PPHOQ from the organized requirements.

Various methods have been applied to determine the fundamental importance ratings of CRs. The simplest method for determining the fundamental importance ratings are based on point scoring scale. Conjoint analysis method was attempted to determine the fundamental importance ratings. Some researchers, for example, Armacost et al., 1994, Lu et al., 1994, described the use of analytic hierarchy process (AHP) for determining the fundamental importance ratings. A group decision-making technique to determine the fundamental importance ratings was proposed by Ho, Lai, and Chang (1999). Owing to the vagueness and imprecision of the importance assessment of CRs, fuzzy set approach or the integrated fuzzy set approaches, for example, triangle fuzzy numbers (Chen et al., 2005, Wang, 1999), fuzzy AHP (Kwong & Bai, 2003), fuzzy arithmetic (Chan & Wu, 2005), fuzzy analytic networks process (Kahraman, Ertay, & Buyukozkan, 2006), were used to determine the fundamental importance ratings. However, the selection of the membership function in fuzzy set is difficult and affected by the subjective experience (Pawlak & Skowron, 2007).

Competitive priority ratings for a company can be obtained by analyzing the company’s relative positions from these customer perceptions, usually using sales point concept. To analysis company performance ratings more objectively and convincingly, the traditional sales point concept was modified or other methods were applied. Chan and Wu (1999) used the entropy method to obtain competitive priority ratings of CRs. Chan and Wu (2005) integrated the improvement rates of CRs into customer competitive analyses. To achieve the total customer satisfaction in an economic way, Kano’s model was incorporated into customer competitive analyses to help accurately and deeply understand the nature of the voice of customer (VOC) (Tan & Shen, 2000).

In summary, most existing methods for determining the final importance ratings of CRs with the uncertain and vague information in PPHOQ depend much on subjective justification and thus lack objectivity. As a result, this may affect the priority analysis of CRs and subsequently lead to an inappropriate decision-making in product improvement.

Rough set theory is a fairly new intelligent tool that is widely used for finding data dependencies, evaluating the importance of the attributes, discovering the pattern of data, reducing all redundant objects and attributes, and seeking the minimum subset of attributes. Moreover, the main advantage of rough set theory is that it does not need any preliminary or additional information about data like a grade of membership or the value of possibility in fuzzy set theory (Pawlak & Skowron, 2007). Therefore, compared with fuzzy set theory, rough set theory can be applicable to a much wider variety of design problems where the determination of CRs in the PPHOQ and their fundamental importance ratings are involved in a vague, uncertain and imperfect way.

The wide applicability of AHP approach is owing to its simplicity, ease of use, and great flexibility. In addition, it can be integrated with scale method in order to consider both qualitative and quantitative factors. The integrated approach can definitely make a more realistic and promising decision than the stand-alone AHP (Ho, 2008). Therefore, this approach is used to determine both the total input and the total feasibility of achieving the improvement ratio of satisfaction estimation (IRSE) of a CR. And then an approximate function is derived for determining the importance rating of achieving the IRSE of a CR.

The rest of the paper is organized as follows. The next section specifies that an approach of acquiring CRs in PPHOQ is proposed with the help of relative reduction method in rough set theory. Section 3 describes that, by using positive region method in rough set theory, a method for determining the fundamental importance ratings is presented. Section 4 describes calculating formulas of the importance rating of achieving the IRSE of a CR. Section 5 specifies the method for determining the final importance ratings of CRs. Section 6 illustrates a case study of a fully automatic dishwasher to empirically verify the feasibility and effectiveness of the proposed approach. The characteristics and limitations of the proposed approach as well as some future research potentials are discussed in Section 7.

Section snippets

Building decision system for acquiring CRs in PPHOQ

A company should accurately know what the customers concerned for the product. This information could be obtained from the company’s sales network or through market survey. Suppose that P customers are selected from all of the company’s customers to help identify CRs for the company’s product. Expressed by customers’ words, Mo original requirements are identified. By using affinity diagram or cluster method, these Mo requirements are categorized into N0 requirements which are higher-level, more

Fundamental importance ratings of CRs

According to the new set of conditional attributes NCcr and the set of decision-making attribute Dcr, Ucr is simplified as the new sample set Ufr. Corresponding to DScr, decision system DSfr = (Ufr, Cfr  Dfr) for determining the fundamental importance ratings of CRs is built, for Cfr = NCcr, Dfr = Dcr.

The fundamental importance rating of each CR in PPHOQ is measured by its contribution to the total customer satisfactory degree, so the fundamental importance rating is defined as the rating of the

Concept of importance rating of achieving the IRSE of a CR

Competitive information can be obtained by asking the customers to rate the relative satisfactory estimation of the company and its competitors on each CR and then aggregating the customers’ ratings. Useful ways of conducting this kind of comparison analysis are also via mailed surveys and individual interviews.

Let us denote the company in question as Co1. Suppose that Q  1 competitors, denoted as Co2, Co3, …, CoQ, are identified. Then the companies’ satisfactory estimations of CRs can be denoted

Final importance ratings of CRs

According to the previous information, the company’s sales point for CRs must be taken into account. A sales point contains such information that characterizes the company’s ability to sell the product based on how well each CR is met. In other words, a sales point indicates the possibility that will give the company a unique selling proposition: when the company and its competitors are all doing poorly at a requirement, one can assume the reason for this is a bottleneck in technology and the

Case study

To demonstrate the performance of the proposed method for rating the importance ratings of CRs in product planning, a case study of product improvement of a fully automatic dishwasher is given in this section.

Conclusions

Determining the final importance ratings of CRs is a fundamental problem in QFD applications. In this paper, a systematic and operational approach is proposed to ascertain the final importance rating of CRs. Firstly, by using some related methods of relative reduction and relative core in rough set theory, a decision system is built and analyzed to determine the CRs of the PPHOQ. Secondly, based on the method of relative positive field in rough set, the decision system is simplified and its

Acknowledgement

This research is financially supported by the National Science Foundation of China (Project Nos. NSFC70721001, NSFC70625001, and NSFC7041028).

References (32)

  • H. Bai et al.

    Inexact genetic algorithm approach to target values setting of engineering requirements in QFD

    International Journal Production Research

    (2003)
  • L.K. Chan et al.

    Rating the importance of customer needs in quality function deployment by fuzzy and entropy methods

    International Journal Production Research

    (1999)
  • Y.Z. Chen et al.

    Fuzzy expected value modeling approach for determining target values of engineering characteristics in QFD

    International Journal Production Research

    (2005)
  • Y.Z. Chen et al.

    Fuzzy regression-based mathematical programming model for quality function deployment

    International Journal Production Research

    (2004)
  • R.Y.K. Fung et al.

    Modeling a quality function deployment planning with resource allocation

    Research in Engineering Design

    (2003)
  • R.Y.K. Fung et al.

    Product design resource optimization using a non-linear fuzzy quality function deployment model

    International Journal Production Research

    (2002)
  • Cited by (0)

    View full text