Some aspects of the analysis of polymer composite materials as complex systems are considered. In this case, the system is presented as a combination of one polymer matrix and several active additives. Within the framework of this work, the composition is assumed to be unchanged, and the properties of the composition are controlled by changing the concentration of the ingredients. In work, a mathematical model was developed that calculates the optimal content of components to improve specific properties of the polymer composition. Obtaining such a model is partly hampered by complex interactions between components, but a solution was obtained within the framework of one specific composition. Nevertheless, it was impossible to transfer these results to other compositions in this case, and it was impossible to obtain a general mathematical model for an arbitrary composition. Therefore, to solve this problem, a black-box model was used in this work. The main methods for studying polymer compositions are presented; their systematization is considered according to the principle of controlling properties at different stages of material synthesis. In this work, a variant of controlling the properties of the polymer composition using active additives was used. The urgency of the problem related to the development of methods for assessing the properties and control of the latter by ranking the concentrations of the ingredients of the polymer matrix has been substantiated. As a result, a mathematical model for optimizing the composition of the polymer composition was obtained. It takes into account the positive and the negative influence of the ingredients on the entire composition of the polymer matrix. Also, computational experiments were carried out to find the optimal concentration of active additives in the composition of the polymer composition under conditions of pair interaction of additives. The model is presented and solved using a quadratic programming problem using a specific example. Different cut-off values were used for the content of the ingredients. The results obtained clearly demonstrate the dependence of the properties of a chemical system on the concentration of specific ingredients. Based on the results of two computational experiments under different boundary conditions, the optimal concentration was calculated for the full manifestation of two properties. The paper also presents a vector of further actions, prospects for improving the model, and possible areas of application of this model.