Study On The Comprehensive Evaluation Technology Based On Interval Information

Comprehensive evaluation has been applied to many fields in the society. As an important branch of economic statistics, comprehensive evaluation are increasingly attracted the attention of society and has become an effective quantitative tools in the assessment, appraisal, and other activities. More and more new ideas, new viewpoints and new methods appear in the comprehensive evaluation, but also exposed the many flaws and shortcomings of their own, all kinds of questions have gradually increased.This situation produced a significant negative impact for the application and its scientific of comprehensive evaluation technology.According to the author's understanding of the factors constraining the scientization of comprehensive evaluation, in addition to the reasons for the institutional level, there is the existence of simplification phenomenon. It refers to the adoption of a single comprehensive evaluation of technical means to one or several steps for processing.The simplification phenomenon in comprehensive evaluation including an evaluation of the subject, the evaluation methods, evaluation of the hierarchy, the evaluation of the time dimension, the evaluation forms and other aspects of the data.In the traditional comprehensive evaluation, the data formats are point value. But different methods have different requirements and regulations to the data structures or evaluation models.In the evaluation process,there will always be qualitative variables, fuzzy variables, interval variables and other data formats.That how to conduct a comprehensive evaluation activities for these different data formats became one of the issues to be studied.The purpose of this article is to expand the point value to interval information in comprehensive evaluation and develope a new evaluation techniques and methods base on interval information.The author believes the comprehensive evaluation technology based on interval information has a wide range of applications and application prospects.On the one hand, from these perspectives,such as the non-comprehensive of data collection,the inadequacy of parameter information,subjective indicators, the ambiguity of weights,the interval number formatin line with the actual situation of comprehensive evaluation. On the other hand, it's more convincing and acceptable to provide a value range of the results of comprehensive evaluation.The author's ideal is to divided the operation process of the interval evaluation into two kinds of methods——Step by Step processing and Black-box processing. In Step by Step processing, according to comprehensive evaluation of the steps, we analyse the disposal processing of data,weight and aggerate method. In Black-box processing,we disscuss the sort of interval value.The chapters are as follows:In Chapter One, author explains the main content of comprehensive evaluation base on interval information, defines the meaning and classification, distinguishes the related concepts, then brings forward the quantification and dimensionless methods of interval indicator.In Chapter Two, the author study the weight of interval number comprehensive evaluation. First, starting from the concept of weight, leads to the feasibility and the basic classification of interval weight. Then,in summarizing the strengths and weaknesses of existed results, the author proposes some method about solving the interval weight base on group decision-making.In order to reflect the differences in group,we designed a indicator to measure the similarity of interval numbers.At last, the author discussed the normalization method of interval weights.In Chapter Three, the author divided the sort of interval numbers into Simpe-interval Sort and Distribution-interval Sort, we proposed Dominance-degree Method for Simpe-interval Sort, sort method based on the cumulative distribution for Distribution-interval Sort.In Chapter Four,we dicussed the aggregate method of interval comprehensive evaluation. The author proposed two ways to aggregate,the one is Transformation way and the other one is Nesting way. In Transformation way,we emphasis how to transformate the interval numbers into point value efficiently. In Nesting way, we discuss some method for interval comprehensive evaluation.In Chapter Five,the author summarized the main research results,meanwhile pointed out the problemsand deficiencies of this paper and further research directions.This paper handles the initial study of the basic problems of interval comprehensive evaluation, and offers the basic ideas of expanding the evaluation data from point value to interval number.Possible innovations of this paper are as follows.(i)The author picks up a new topic. He successfully sets up the basic content of interval comprehensive evaluation and puts forward a series ideas.(ii) Put forward the concept of similarity of interval numbers and design indicator to measure it.(iii) proposed Dominance-degree Method for Simpe-interval Sort.(iv) proposed a sort method based on the cumulative distribution.