An improved Fuzzy Comprehensive evaluation method using expanded least deviations algorithm

Fuzzy Comprehensive evaluation is usually influenced significantly by the matrix of fuzzy relation and weight vector. For a sequential segmentation category, the principle of the lowest cost, the principle of maximum degree of measure and the principle of maximum degree of membership sometimes can get unreasonable conclusion, because they conceal the difference of two degree of membership. First of all, a new expanded least deviations algorithm is presented for combining index weights, then bring out a improved fuzzy Comprehensive evaluation method based on reliability code. The proposed method can overcome the shortages of the traditional fuzzy Comprehensive evaluation. Case results clearly show that the proposed method is attractive and effective.