Plastic, an integral part of modern life, is widely used in various sectors such as automotive, aerospace, and healthcare. The rapid advancements in the plastic industry have improved plastic processing technologies. Among contemporary production methods, plastic injection molding has become one of the most commonly used techniques. As industrial markets evolve rapidly, the need to shorten product cycle times, reduce production costs, and increase production speeds to respond swiftly to demand has become increasingly urgent. In this context, the thesis addresses the reduction of cycle times through the optimization of process parameters in the injection molding process. By utilizing experimental data available in the literature, a mathematical model of the injection molding process has been developed using a hybrid method known as Neuroregression approach and cross-validation technique. To minimize the cycle time of the injection molding process, multi-objective optimization scenarios were created using seven different process parameters and two parameters affecting product quality. Optimization studies were carried out using stochastic optimization methods with the "Simulated Annealing," "Random Search," "Nelder-Mead," and "Differential Evolution" algorithms in the "Wolfram Mathematica" program with the help of the "NMinimize" tool. When comparing the obtained optimization results with those in the literature, it was found that the model and optimization methods used in the study are reliable and applicable.