Affiliations: Department of Industrial, Welding and Systems
Engineering, The Ohio State University, Columbus, OH 43210, U.S.A.
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.
