Optimization of Chinese coal-fired power plants for cleaner production using Bayesian network

Abstract

Concern over the influence of cleaner production (CP) transformation on the industrial enterprises has increased in recent years. This research evaluated each process efficiency that enabled entrepreneurs to choose optional ones that previously executed the whole procedure in CP transformation in industry. The risk of invest (time and cost) for each step of CP transformation was ranked using Bayesian network. Datasets from case study of coal fired power plants were simulated using a method of procedure simulation, Design Structure Matrix (DSM), which sought to be prepared for Bayesian network learning. The procedure simulation was based on relationships that each CP transformation step has with each other achieved from experts and experiences. The research gave theoretical guidance and an operable method to industrial entrepreneurs for arranging CP transformation steps and allow them to separate the whole procedure into several stages if inconsistent with budget. Findings indicate that the science and technology categories are preferred for high efficiency in the coal fired power plant CP transformation case.