The Research On Multi-attribute Comprehensive Evaluation Methods With Incomplete And Uncertainty Information

To a system and its act or feature, people usually can’t have enough quantitative or qualitative information to make complete and certain description, prescribe and forcast. The reasons are cognitive biases, information asymmetry and so on. Therefore, people’s preferences information is often with uncertainty (with the interval numbers, triangular fuzzy numbers and language terms) or incompleteness (the incompleteness of evaluation parameters, such as attribute weights, attribute values and experts weights and so on). Here, for the comprehensive evaluation problem whose evaluation parameters are incompleted and preferences are uncertain, this perpor defines this problem as incompleted and untercain comprehensive evaluation question.Incompleted and untercain comprehensive evaluation is the prolongation and development of the traditional comprehensive evaluation theory. It is an important component of the traditional comprehensive evaluation theory. And the research of this question has important theoretical and practical significance. In theory, using the knowledge of some corses such as scientific management, logistics and optimize metra and so on, people do research on the incompleted and untercain comprehensive evaluation problem can perfect the traditional comprehensive evaluation theory and method; In practice, these methods can be widely used in the fields of economic, management and statistical science and so on.Under the aspects of "incompleted information" and "uncertain information", this paper carries on the following research through the comprehensive evaluation theory and methods.(1) Carry on customary status quo about incompleted and untercain comprehensive evaluation, analyse the collected documents’research results, which mainly contains the problems and weaknesses. And then the paper points out the research strating point.(2) To the situation that the attribute weights and attribute values are unknowed and the evaluator can only determine the sequence of the objects under the same attribute, this paper presents one low-cost and facilitated evaluation method. Besides, the paper presents an improved model about the incompleted and untercain comprehensive evaluation. For multi-attribute comprehensive evaluation problems with segmental evaluation information, this dissertation analyzes the appropriateness of models in the existing literatures and then proposes an improved model with Dempster-Shafer theory. The resolution method, decision-making rule and excellent characteristics of this improved model are then studied.(3) To two types of incompleted information, this dissertation do research on group attribute comprehensive evaluation methods. First, the paper presents a concentrated method for groups of experts weights based on relative entropy. Second, it presents an interaction method for groups attribute decision-making under incomplete information.(4) Based on the traditional sequence relationship analysis, this dissertation presents another sequence relationship analysis method in the form of interval numbers. At the same time, when the attribute weights and the attribute values are both given in the form of interval numbers, this dissertation presents an interval numbers concentrated method based on comprehensive weights. Under the situation of the preference information are uncertain, this paper gives one evaluation method based on the principle of maximizing the preference consistency degree of evaluation objects and attributes.(5) This dissertation does research on group attribute comprehensive evaluation methods to two types of incompleted information. When the evaluator weights are known while the attribute weights are unknown, this dissertation presents one mixed attribute group evaluation method. And when the evaluator weights are unknown while the attribute weights are known, the dissertation presents another mixed attribute group evaluation method...