Robust and reliability-based multidisciplinary optimization of injection molding system

Abstract

This thesis focuses on multidisciplinary design optimization (MDO) of amulti-cavity injection mold insert taking into account the multidisciplinary nature of the injection molding system. The proposed modifications in the conventional multi-cavity insert will allow flexibility in the production of a specified plastic part family. The use of the proposed flexible multi-cavity injection mold insert can contribute to the reduction of overall mold manufacturing costs. Aninjection molding machine using flexible multi-cavity inserts need to mold different part family members, whose specifications vary within a permissible limit. Hence, there exists uncertainty in the input variables. Two approaches: robustness-based design optimization (RDO) and reliability-based design optimization (RBDO) are used to handle the uncertainty in the input variables.The optimal design for the proposed multi-cavity insert considers the minimization of cycle time and pressure drop simultaneously. Therefore, a multi-objective optimization method is used in the problem formulation. Thus, this thesis solves the design problem using two formulations: the first oneintegrates MDO, RDO, and multi-objective optimization; the second oneintegrates MDO, RBDO, and multi-objective optimization.A numerical illustrationof the design problem is presented specifying a part family with respect to a particular model of injection molding machine and a particular polymermaterial. Finally, Pareto optimal solutions are obtained includingalternative combinations of cycle time and pressure drop. The flexiblemulti-cavity insert can be manufactured using the optimal values of the design variables corresponding to the selected combination of cycle time and pressure drop. The proposed moldinsert will be aneconomical solution forthe production of small plasticparts having dimensional similarity,e.g.,components of toys, cosmetic boxes, household containers, electric switches, etc.