Systematic waste minimization in chemical processes. 2. Intelligent decision support system

Abstract

In part 1 of this paper, a systematic methodology for a waste minimization analysis applicable to any chemical processes is proposed. In part 2, we present the architecture of ENVOPExpert, an expert system that implements the methodology. Given information concerning the process in the form of a flow sheet, process chemistry, and material information, ENVOPExpert can automatically detect the waste components in the process, diagnose the sources where they originate, and suggest intelligent design alternatives to eliminate or minimize them. ENVOPExpert consists of waste minimization domain knowledge organized into process-general waste minimization knowledge and process-specific information. The inference engine contains algorithms for qualitative simulation, waste diagnosis, and alternatives generation, which use the process-specific information as a basis to perform the analysis. ENVOPExpert also assists the user to screen and rank the various alternatives using a waste minimization index. The application of ENVOPExpert to an industrial case study is illustrated, and the results obtained are compared to the available experts' solutions.