Neuroevolutionary Multiobjective Methodology for the Optimization of the Injection Blow Molding Process

Abstract

Injection blow molding process is widely used in the industry to produce plastic parts. One of the main challenges in optimizing this process is to find the best manufacturing thickness profiles which provides the desirable mechanical properties to the final part with minimal material usage. This paper proposes a methodology based on a neuroevolutionary approach to optimize this process. This approach focuses on finding the optimal thickness distribution for a given blow molded product as a function of its geometry. Neural networks are used to represent thickness distributions and an evolutionary multiobjective optimization algorithm is applied to evolve neural networks in order to find the best solutions, i.e., to obtain the best trade-off between material usage and mechanical properties. Each solution is evaluated through finite element analysis simulation considering the design of an industrial bottle. The results showed that the proposed technique was able to find good solutions where the material was distributed along the most critical regions to maintain adequate mechanical properties. This approach is general and can also be applied to different geometries.