Multi-attribute Decision Theory, Method And Its Application In Mining Research

There are many multi-attribute decision-making problems in the fields of society, economy and management, such as investment decision-making, project assessment, quality assessment, alternative selection, plant site selection, resources assignment, evaluation of scientific research achievements, personnel appraisal, ranking of industrial departments and comprehensive evaluation of economic benefit, etc. Therefore, the theories and methods of multi-attribute decision-making show a prospect of wide range of applications. It has great significance in theory and in practice to make systematic study on the theories and methods as well as their applications.In this thesis, many kinds of knowledge are used. They are system science, management science, operations research, behavior sciences, psychology, probability and statistics, fuzzy mathematics, grey system theory and mineral industry economics. The research work of this thesis Consists of two parts, i.e. the systematic study on the theories and methods of multi-attribute decision-making and their applications in mineral industry.Main research contents of the theories and methods of multi-attribute decision-making are outlined as follows:Determination of attribute values under qualitative attributes is subjective and fuzzy. In this dissertation, the theory of fuzzy set and the fuzzy judgment matrix are used to deal with the quantitative problem of qualitative attribute in order that the decision-making matrix is determined in scientific and objective ways.The goal of formalizing the decision-making matrix is to eliminate the influence of dimension on decision-making result. In this dissertation, the author analyzes several methods of formalizing decision-making matrix, points out their merits, defects and application areas, and puts forward a new method to formalize decision-making matrix, named transform formula based on exponential function.Considering the defects of the present subjective and objective weighting method, the author proposes two kinds of new combination weighting method, one is based on the essence of attribute weight, the other is based on the decision-maker's preference to the weighting method and consistency degree of several weighting methods. These two methods are more scientific and reasonable than other comprehensive analysis methods in determining attribute weight.Study on the theory and method of combination decision-making is almost a blank in decision domain. In this thesis, the author studies on combination decision-making method of multi-attribute decision-making and presents a kind of combination decision-making method based on minimizing the square sum of deviation. Furthermore, the disadvantage of the combination decision-making method based on desirability vector is pointed out. This work lays a foundation for general research on combination decision-making.As respect to comprehensive ranking of alternatives, this thesis presents a multi-attribute decision-making method based on the idea of combination weighting and Ideal Point Method. A practical example shows that this method is reasonable and feasible, and has a good discrimination effect.The theory and method of multi-attribute group decision-making are studied in this dissertation. (1) Four types of multi-attribute group decision-making are given. (2) The concept of objective weight of decision-makers is put forward, and the method to determine the objective weight ofdecision-makers on the base of AHP judgment and ranking vector is given respectively. Moreover, the subjective and objective weight are integrated into final weight of the decision-maker by means of convex combination. (3) Two group ranking methods based on priority rank of alternatives are given, one is the model of minimizing the weighted sum of deviation square and the other is the model of 0-1 programming based on measure function. In addition, considering the weight of the decision-making, the author generalizes the Fuzzy Borda Method to the case of multi-attribute group de...