Sink-mark minimization in injection molding through response surface regression modeling and genetic algorithm

This paper deals with minimization of sink depths in injection-molded thermoplastic components by integrating finite element (FE) flow analysis with central composite design (CCD) of experiments and genetic algorithm (GA). Sink-mark depth depends on various process and design variables. Out of all, four most influential variables viz. melt temperature, mold temperature, pack pressure, and rib-to-wall ratio were used for optimization. A set of FE analyses were conducted at various combinations of variables based on the CCD array. A second-order-response surface regression model (RSRM) was developed based on the CCD. The second-order model was effectively coupled with GA for optimization of variables to minimize the sink depth. Results are encouraging and the proposed methodology could be used effectively in minimizing sink-mark depths.