The automation of control in continuous chemical and technological processes necessitates the use of numerical methods capable of accounting for complex reaction kinetics and multiple interdependencies among process parameters. This study proposes an approach to automated selection of control actions based on the integration of mathematical modeling and heuristic optimization using a genetic algorithm. The mathematical model is formulated as a system of ordinary differential equations describing the dynamics of reactants and products in a cascade of ideal mixing reactors. Aggregated indicators are used to evaluate the properties of the final product, in particular the number-average and weight-average molecular weights, which are computed via statistical moments of the molecular weight distribution. The genetic algorithm performs a global search in the space of control variables and allows identification of component feed strategies that ensure achievement of target technological characteristics. The proposed approach has been practically implemented and validated through an industrial-scale case study of butadiene-styrene copolymerization, where the dosing strategy of a molecular weight regulator was optimized. The results demonstrate high accuracy in reproducing the specified product parameters and confirm the applicability of the method for intelligent process control. The developed approach can be integrated into modern digital manufacturing systems, including digital twins and adaptive control loops in the chemical industry and related sectors.