The injection molds consume costs and time for manufacture. Therefore, it is important to predict the molding method. The CAE technology can analyze the resin flow in designed injection molds. This method is useful for checking the validity of a design. However, this technology cannot make the indication or the automatic correction of a point that improve design variables, such as the runner diameter, and the amount of resin, when fault arises in initial setting. Therefore, a design change must be made by the trial and error of the engineer.
In this research, the genetic algorithm (GA) is improved so that the multi-cavity molding of an injection molding can be automatically determined using the GA, which is one of the discrete optimization method.