LAUSR

Abstract Environmental impact evaluations in manufacturing fields is a complicated decision-making problem, which involves wastes emission, resource consumption and energy utility, etc. This paper presents an intelligence method for environmental impact evaluation using kernel fuzzy clustering and a back-propagation neural network. The objective of this article is to apply artificial intelligence technology to evaluate environmental influence of cutting process and develop a decision support tool for selecting the optimal environmental solution from various alternative schemes. There are three stages to accomplish this assessment procedure. First, the evaluation index system of the cutting process was analyzed and established. Next, a training sample set which was used for learning the ‘knowledge and experience’ by neural networks, was acquired by kernel fuzzy clustering algorithm. Finally, an evaluation model based on a back-propagation neural network was constructed, and connection weights of the model were determined after training. A case research on the cutting processes of the automobile workpiece was conducted. Results showed that the proposed method is more concise and practicable than existing evaluation methods and provide a feasible and effective decision-making tool for selecting an optimal cutting process while manufacturing.