Optimization of injection molded parts by using ANN-PSO approach

Purpose: The aim of the work was the optimization of injection molded product warpage by using an integrated environment. Design/methodology/approach: The approach implemented took advantages of the Finite Element (FE) Analysis to simulate component fabrication and investigate the main causes of defects. A FE model was initially designed and then reinforced by integrating Artificial Neural Network to predict main filling and packing results and Particle Swarm Approach to optimize injection molding process parameters automatically. Findings: This research has confirmed that the evaluation of the FE simulation results through the Artificial Neural Network system was an efficient method for the assessment of the influence of process parameter variation on part manufacturability, suggesting possible adjustments to improve part quality. Research limitations/implications: Future researches will be addressed to the extension of analysis to large thin components and different classes of materials with the aim to improve the proposed approach. Originality/value: The originality of the work was related to the possibility of analyzing component fabrication at the design stage and use results in the manufacturing stage. In this way, design, fabrication and process control were strictly links.