Research On Multiple Attribute Decision Making Method Based On Interval Rough Numbers

Multipe attribute decision-making (MADM), which is one of the most important parts of decision-making theory, is widely used in many fileds, such as project evaluation, economy management, engineering ect. Objective things are complex and changeable, human thinking is fuzzy, so decision-making information is often given in the form of interval rough numbers in practical decision making problems. Now, for this kind of problem research is still relatively little. It has great signifycance for sovling practical problems that making depth and systematical study of MADM methods which the attribute values are the interval rough numbers.This paper focuses on the attribute values are interval rough number multiple attribute decision makingproblems for research, the main work is as follows:In chapter1Euclidean distance and the expected value of interval rough numbers are defined. Give new L-measure of the interval rough numbers, by which interval rough number can be sorted on the new measure.In chapter2It defined a concept of connection number formula for the interval rough number and gived apossibility degree formula for the comparison between two interval rough numbers.In chapter3A decision making method is proposed based on fractional programming for the problem of multiple attribute decision making, in which the attribute values are the interval rough numbers and the weights of attributes are intervals. By using the TOPSIS, the models of non-linear fractional programming for alterative’s relative closeness are built. The alternative’s relative closeness are obtained for ranking all alteratives.In chapter4A decision making method is proposed based on connection number and comentropy for the problem of multiple attribute decision making, in which the attribute values are the interval rough numbers and the weights of attributes are unknown. A decision making method is proposed based on apossibility degree for the problem of multiple attribute decision making, in which the attribute values are the interval rough numbers and the weights of attributes are real number.These decision making methods are new multiple attribute decision makings, in which the attribute values are the interval rough numbers.Finally, the works of this paper are summarized and some expectations in future are proposed.