Taguchi‟s method is a powerful problem solving for improving quality of the product and its basic principle which examine the effects of many process variables and identify those factors which have major effects on process using a few experiments. The parameters including are melting temperature, injection pressure and cooling time were selected in plastic injection molding process. Due to these process parameters different shrinkages defects are happened in the injection molding process of the inside and outside dimension of plastic beverage crate products. But the companies are trying to minimize shrinkage without any optimization tech-niques with trial and errors; as a result, they didn‟t minimize shrinkage and lacks effective-ness. For the research work plastic beverage crate, made from high density polyethylene (HDPE) material using injection molding process were selected. The numbers of experiments conducted were based on the number of parameters and levels with three levels and three factors of L9 Orthogonal Array (OA). The levels have been se-lected based on the standards acceptable and plastic manufacturer experience in this organiza-tion for this plastic factory. In this study optimal plastic injection molding process parameters were determined using S/N ratio and ANOVA methods that minimizes shrinkage with “small-er the better” S/N ratio approach with the application of Minitab 15 software. According to the result optimum parameter values are 290 ℃ of melting temperature, 130 MPa injection pressure and 60 second cooling time were found. From analysis of variance result, melting temperature was the most significant factor (15 %), cooling time is the second of (12 %) and injection pressure was the least significant (9 %) for plastic beverage crate of the injection molding process. The optimum response variables (shrinkage) were found as 6.03mm in the mean S/N ratio response analysis and the conformation experimental test gave 6.11mm, which is approximate to the optimal mean S/N ratio of the response result and it was minimum as compared with the other eight experimental runs and the existing shrinkage val-ues (6.81 mm) with a margin of error (1.3 %).The mean response prediction ranges for future work were estimated at 5.85 < 6.309 < 6.75 mm.