Dynamic Comprehensive Evaluation Method With Fuzzy Rewards And Punishments

Having broad application prospects, comprehensive evaluation is objective, fair and reasonable overall assessment made to the evaluation objects, which could be found abroad in the fields of sociology, management, economics and culture. With the introduction of the time dimension, static comprehensive evaluation become into a dynamic comprehensive evaluation.Dynamic comprehensive evaluation is an important research direction of the comprehensive evaluation field. The essence of the dynamic comprehensive evaluation problem is to reduce dimension.Existing research has focused on the process of reduce the dimension and determining the weights in the different time period.Multiple phase information aggregation in dynamic comprehensive evaluation is a process that gathering the value of evaluation objects or systems at some times to a point value. Existing research for multi-stage information assembled includes the assembly method based on the time factor, the build-up method of the difference in the driving principle and the assembly methods of the perspective of incentives or penalties,etc.This paper suggests an aggregation method with fuzzy rewards and punishments is suggested. The main research works of this paper are as follows:(1) Multiple phase information aggregation method is theoretically perfect. On the basis of existing research on multiple phase information aggregation method, a new method with fuzzy rewards and punishments have been defined by using the knowledge of fuzzy mathematics to make the boundaries unclear. The corresponding information aggregation model has been put forward to implement fluctuant rewards or punishments on evaluation objects.(2) Corresponding improvement has been put forward with the consideration of "value of time". In order to verify the validity of the method, the same example was applied to different methods, and the results were analyzed and compared briefly.(3) Considering the effect of subjective factors, under the determined membership function, the influence of no rewards and punishments point to evaluation results were analyzed.