Probability numbers for multi-attribute decision-making
Shizhou Weng,Zhengwei Huang,Yuejin Lv
DOI: https://doi.org/10.3233/jifs-223565
2024-01-17
Journal of Intelligent & Fuzzy Systems
Abstract:In the face of increasingly complex data forms and decision-making problems, the uncertainty of information poses a major challenge to multi-attribute decision-making methods. How to effectively organize information and serve realistic decision-making problems has attracted extensive attention in the academic circles. In view of this, based on the distribution law of random variables, we put forward the basic concept of probability numbers and construct a general framework, including the concepts of type, order, item, isomorphism and isomerism, same domain and same distribution of probability numbers. On this basis, we further define the expectation and variance formula of probability numbers, and its operation rules are defined for the same type of probability numbers. To compare the dominance and inferiority of probability numbers further accurately, we put forward the concepts of dominance degree and comparability degree of probability numbers, so that decision makers can realize the ranking of probability numbers by calculating the comprehensive dominance degree. In view of the related concepts of probability numbers, we summarize the properties and theorems of probability numbers and prove them. In addition, a probability numbers-based multi-attribute decision-making framework model is proposed to solve the multi-attribute decision-making problem. Decision makers can select appropriate sub-models to construct personalized multi-attribute decision-making methods according to actual needs. At the end of the paper, we apply the method to the multi-attribute decision case of campus express stations evaluation and verify the scientificity and rationality of the evaluation method. The concept of probability numbers and its decision model proposed in this paper extend the concept category of numbers, enrich the multi-attribute decision-making method based on probability numbers, and have certain reference significance for further research of uncertain decision theory and method.