Abstract: Defects emerge more easily in thin wall injection molding because of the higher injection pressure, higher injection speed, and higher melt temperature. In this case, it is necessary to study process condition by simulation and control the defects. Injection molding defects are mainly researched by applying the theory of artificial intelligent and vague theory. The diagram of vague expert system in product quality control and the inference flow diagram of the system were given in the paper. The membership function of process parameters and defects and the fuzzy rules were constructed. The membership function reflected the fault vague and vague connection. Vague reasoning based on the rules was realized. A case study was carried out to test the defects optimization module.

Key words: thin wall injection molding technology, expert system, membership function, fuzzy inference