Injection molding is an established technology with many decades of use; however, factors such as the processed raw material, the mold and machine, and the processing parameters result in significant variations in product qualities. Traditionally, researchers have attempted to improve the quality of injection-molded components by controlling screw position, injection and holding pressures, and mold and barrel temperatures. However, the geometrical dimensions of molded components still tend to vary from shot to shot even when such controls are applied. In other words, a further work is required to properly understand the factors affecting melt quality in each cycle such that more effective control strategies can be put in place. Conventionally injection molding is a ‘black box’ and thus process parameters setting is often time consuming and has limited effect basing on statistically experimental methods, computer aided simulations, or operator’s experiences. With the advances of sensing technology, understanding of injection molding process is transformed into a ‘grey box’ that reveals the physical information about molten resin’s flowing behaviors in the cavity. Thereby, this study develops an intelligent optimal parameters tuning approach that analyzes the profiles of injection speed, injection pressure, cavity pressure, screw position of injection molding, and the process parameters setting data provided from the controller of injection molding machines. Further, the major parameters at each stage are determined individually including injection speed/pressure in the mold filling stage, velocity-to-pressure switchover point, holding pressure, and packing time. Through injection molding of an IC tray as an example for experimental verification, this study presents the initial results of developing an intelligent computing approach that determines the above-mentioned process parameter optimization. In addition, the molding curves can be used for process monitoring.