Research On Application Of Fuzzy Analytic Hierarchy Process-Fuzzy Comprehensive Evaluation Model

In the investigation of the effectiveness of volunteer teaching practice,this paper finds that the traditional statistical methods have some shortcomings,such as ambiguity and unclear purpose.Therefore,this paper combined the fuzzy analytic hierarchy Process（FAHP）and fuzzy comprehensive evaluation（FCE）to build a fuzzy analytic hierarchy process-Fuzzy comprehensive evaluation（FAHP-FCE）model to analyze the effectiveness of volunteer teaching practice.In practice,FAHP-FCE model also has some problems,such as large scale and subjective decision.Then the improved model is applied to the performance evaluation of government livelihood projects,and good results have been achieved.The specific research work is as follows:（1）This paper first uses traditional statistical analysis methods to investigate and analyze the effectiveness of volunteer teaching practice for normal university students in poverty alleviation work.In this part,SPSS was used to conduct factor analysis on the index system.The 23 factor loads were all up to the standard,and the intra-group differences showed by descriptive statistical analysis were not obvious.Then,independent T-test and variance analysis were conducted on relevant background variables,and it was found that there were large or small differences among different respondents,but the importance of the impact factor itself could not be determined,and comprehensive evaluation could not be carried out.Therefore,the FAHP-FCE evaluation model is used to conduct a secondary analysis on the effectiveness of volunteer teaching practice.（2）In the application of FAHP-FCE model on the effectiveness of volunteer teaching practice for normal university students,the indicator system used is still derived from previous investigations.Based on the characteristics of fuzzy analytic hierarchy process（AHP）,the factors are sorted and stratified to obtain a reasonable index set and construct the AHP structure.The comparison of indicators is made by experts,and the order of importance of all indicators can be obtained through the judgment matrix,so that the influence degree of factors and the difference of importance between each factor can be seen concisely and clearly.Then,based on the fuzzy comprehensive evaluation method,five evaluation grades were divided,and the effectiveness of this volunteer teaching practice was comprehensively evaluated based on the survey data.In this evaluation process,due to the primitive scale of the importance degree of evaluation indicators,and the subjective decision-making of experts,it is easy to lead to cognitive bias,narrow knowledge and other errors,and there may be a gap between the evaluation results and the reality.The scaling and expert decision are improved.（3）For the problems of scale and expert subjectivity,the concepts of scale table and group decision making combined with 1-9 scale and AAA scale are introduced respectively.The original 1-9 scales were evenly spaced,too rough in describing the degree of importance,and insensitive in the case of fine index differences.The new scale,which combined 1-9scales with AAA scales,was more suitable in the case of complex and large index system.The idea of group decision making is to change a single expert into many experts and replace individual decision making with group decision making to make the judgment matrix more scientific and objective.The two improved methods are applied to the performance evaluation of government livelihood projects,and combined with the government self-evaluation,the evaluation is carried out from multi-dimensional and multi-dimensional perspectives.As the data applied to the model is usually only applied once,it is not tested through a series of time series data,so it is impossible to judge whether the model is only valid in this instance.Later,the concept of historical confidence will be introduced to further optimize the model.