Abstract
Injection molding (IM) is considered to be one of the most important mass production processes for plastic products. A substantial amount of research has been directed towards finding settings for the IM process variables as well as the optimal location of the injection gates. These objectives have been mostly approached through the optimization of performance measures (PMs) as functions of the process' variables. Several times, however, these process design and part or tooling design objectives have been addressed independently, rather than in an integrated fashion. The use of computer-aided engineering (CAE) has played a pivotal role in trying to achieve these objectives. The aim of this work is to demonstrate a method based on CAE, statistical testing, artificial neural networks (ANNs), and data envelopment analysis (DEA) to find the optimal compromises between multiple PMs to prescribe the settings of IM process variables and the location of the injection gate. A case study involving the production of a generic part with cut-outs similar to window frames is presented for this purpose. The PMs in this case include the part flatness, the maximum injection pressure, and the time the molded part takes to solidify. Our results demonstrate why this optimization project could have not been properly carried out by taking each performance measure independently.