Weights generation models based on acceptance degrees in decision making

Abstract

The process of determining weights for a collection of experts is an essential component in addressing collective decision-making issues. In cases where individual evaluation values are accompanied by uncertainties, it is feasible for each expert to endorse the evaluations of their peers without necessitating further interaction among the group. This study proposes innovative approaches to determining weights, primarily relying on measurements of the overall acceptance degree. Additionally, the guidelines for computing the total acceptance degrees are established. Various methods can be employed to derive and calculate the total acceptance degree of an expert based on the evaluations provided by other experts. The study at hand introduces several novel concepts, namely “parameterized family of uncertainty functions” and “uncertain system”, which can be effectively utilized for the development of relevant algorithms. The mathematical properties pertaining to the proposed concepts have been scrutinized and subsequently expounded upon. A normalized weight vector can be derived directly from any vector of the obtained total acceptance degrees. Numerical examples have been provided to serve the purpose of illustration and comparison.