Multi-attribute decision making: An innovative method based on the dynamic credibility of experts

Multi-attribute decision making has become a topic of interest for scholars because it can comprehensively and effectively be used to make decisions in situations in which there are multiple homogeneous options. Attribute weighting is an important step and has a significant impact on decision-making, and the subjective weighting method is commonly used in reality. However, as experts have different knowledge, experiences, preferences and so on, the weights of attributes given by experts are subjective. So expert credibility affects the final weights, and the correctness of the weights calculated in this case cannot be guaranteed. Therefore, the dynamic expert credibility model (DECM) is proposed. First, based on the decision matrix and the weight evaluation matrix, the method for calculating distance-based expert credibility calculates the distance between expert evaluations via the score deviation and ranking deviation. Second, considering the differences in the weight evaluation matrix caused by changes in the individual background of the experts, the expert background change process (EBCP) is proposed. Third, the dynamic value of credibility before and after the EBCP can be calculated. To prove the validity of the model, a test method is proposed from the perspective of data envelopment analysis. Finally, evaluations on industrial economic benefits of 16 provinces or municipalities in China are conducted to illustrate the applicability of the proposed model in practice. Using the test, DECM effectively eliminates the influence of the weight calculation due to expert credibility. After the EBCP, the target value of the DECM reaches 6.0989 and the validity of attribute weights is improved by 2.30%. Compared with the traditional weight determination method, the decision-making result under the DECM is consistent. (C) 2020 Elsevier Inc. All rights reserved.

Automated summary

The dynamic expert credibility model (DECM) proposed in this study aims to address the issue of weight calculation affected by expert credibility. Through model validation, DECM significantly improves the effectiveness of attribute weights and effectively eliminates the impact of expert credibility on decision-making results.