Abstract: The energy consumption of discrete manufacturing system presents many complex characteristics such as multi-source, multi-level and dynamic uncertainty, which make energy efficiency analysis and evaluation very difficult. The evaluation index system of energy efficiency of discrete manufacturing system is constructed from three aspects: manufacturing equipment, processing tasks and products. The improved Bayesian probability evaluation model is adopted for energy efficiency evaluation. In determining the weight link, the dynamic modified AHM method is combined with entropy weight method to make up the uncertainty of the structure and parameters of the Bayesian model, and to synthesize the subjective and objective two aspects of the index weight value, and to improve the stability and objectivity of the energy efficiency evaluation algorithm. An example is given to illustrate the effectiveness of the method.

Key words: manufacturing system, evaluation index system, Bayesian method, synthetical evaluation