Study on Bayesian Congeries and Simulation Approach in Stochastic Multi-attribute Decision Making

摘要

Combining decision makers' attitude about risk with stochastic attributes and their stochastic weights of schemes should get better selection in stochastic multi-attribute decision making. But it is real difficult in studying this problem. A method is proposed based on Bayesian theory and statistical simulation in this paper. A multivariable normal congeries model is built to get identical probability distribution of each attribute value. By means of statistical simulation, weights of each attribute are created. By calculating the rank probability of scheme and the comprehensive rank value of scheme which embodying decision makers' attitude about risk, a rank of schemes is got. Finally, an empirical analysis shows the effectiveness of this method.