On the expert system by means of applying principal component analysis for cocoon cooking

Abstract

Recently, filature factories have many problems in daily work because of decrease of the cocoon cooking experts. Irregularity in size and quality of cocoon, which is caused by decrease the cocoon production in each sericultural farm household, accompanied by great difficulty in cocoon cooking technique. In order to solve these problems, the authors introduced an expert system by means of applying principal component analysis into the cocoon cooking process. This system is a computer aided consulting one in the process, and contains theoretical and experienced knowledge of experts in the cocoon cooking. Namely, the system infers the correct prescription by using the theoretical and experienced knowledge of experts, and solves the technical problems instead of the cocoon cooking experts. In this paper, this system was applied to a filature factory, and an examination was made of the relationship between the reeling results and the cocoon cooking technique. As a result, it was confirmed that the expert system was useful for supporting unskilled engineers in the filature factory.