Applying Fuzzy Multi-criteria Decision Method to Evaluate the Credibility Ranking for Taiwan Banks' potential Debtors from the Viewpoint of Default Probability - A Taiwan LCD Panels Industry Case

The purpose of this paper is to provide an algorithm for commercial banks in Taiwan to evaluate the credibility of their potential debtors from the default probability viewpoint under fuzzy environment. The fundamental concepts we have adopted include the eigenvector method, fuzzy Delphi method, fuzzy set theory, and multi-criteria decision making method. By coordinating the concept of internal credit rating approach proposed by Basel II and the interviewing senior managers of bank industry in Taiwan, we propose four major criteria and forty sub-criteria to construct the hierarchical structure for evaluating the banks’ potential debtors’ credibility. The fuzzy Delphi method is integrated with the eigenvector method to form a set of pooled weights of the criteria. The concepts of triangular fuzzy number and linguistic variables are used to assess the preference ratings, including ‘importance’ and ‘appropriateness’, of linguistic variable. Through the hierarchy integration, we obtain the final scores of credibility, and then use a revised Chang and Chen’s ranking method for evaluating the ranking of debtors’ credibility.