The Uncertainty Decision Making Method And Its Application Of Things With Multi-indexs And Multi-levels

Decision making is in all areas of human life, and persist throughout the whole process of human life, human life in some sense is a decision process. People often facing with problems are complex, to reflect these problems, a number of features needed to extract, not only through the things a feature to reflect, each feature we extracted may contain several sub system, which is a multi-level and multi-attribute problem, this kind of problem widely exists in the social, economic, management in various fields, such as investment decision, project evaluation, optimization, water quality evaluation, local economic strength ranking and other aspects. These decision object includes a plurality of indicators, each indicator is composed of several sub-systems. Multiple index, multilevel decision making is to describe how the evaluated things different aspects and dimensions of different indicators, into dimensionless relative values, and these values in order to draw on the affairs of a whole evaluation system. But in reality, the decision information is not fully determined, such as the index weight information, index information, preference information, this uncertainty exists in the subjective and objective measurements, and different dimensions to compare, it is not easy to identify. Therefore the establishment of multiple indicators, multiple levels of decision making under uncertainty evaluation method model, discusses its application in actual problem, is not only a theoretical problem, but a significant application prospects of the practical problems. In this paper, the integrated use of management science and engineering, fuzzy mathematics, linear programming, systems engineering theory, analyses such as analytic hierarchy process, fuzzy comprehensive evaluation, clustering of rough set algorithm of existing typical evaluation method. Based on the existing results to put forward new ideas, fuzzy mathematics fuzzy information is the crucial step to realize the conversion of membership degree, namely from the single index membership degree to target in multi index objective membership conversion. Hierarchy analysis method in hierarchical structure in the target on the bottom ranking is the key step in order to achieve this transformation, the current two conversion methods are based on the linear method. However, the above two kinds of conversion is not linear but nonlinear conversion conversion, this paper gives the fuzzy membership degree of conversion and AHP sequence conversion is nonlinear conversion method. According to the analytic hierarchy process and fuzzy mathematics in two unique nonlinear transformation, targeted research to achieve the conversion of a class of nonlinear calculation method, this method contributes to solving a large class of practical problems, so it has important value in theory and application, combined example to show this method is effective and superior.