Injection molding (IM) is the most important process for mass-producing of plastic products. The difficulty of optimizing an IM process is that the performance measures (PMs) usually show conflicting behavior. The aim of this work is to demonstrate a method utilizing CAE, statistical testing, artificial neural networks (ANNs), and data envelopment analysis (DEA) to find the best compromises between multiple PMs, considering the variability in these PMs in an explicit manner. Two case studies are presented. The first case study, based on a virtual part, is discussed in detail in order to illustrate this method. The second case study is experimentally based and makes use of the American Society of Testing Materials (ASTM) mold to illustrate how this approach applies when purely experimental results are available.