The article describes the implementation of an evolutionary-simulation algorithm in the simulation system of the metallurgical enterprises processes. The algorithm is based on the multi-agent genetic optimization method proposed by the authors in previous works. The multi-agent genetic optimization method is intended to solve optimization problems where the objective function is not formalized analytically. The evolutionary-simulation algorithm has been implemented on Java in the optimization module of the metallurgical enterprise information system. The algorithm has been tested in solving the problem of optimization of the logistics processes of the steel products manufacture. During algorithm testing, recommendations have been obtained for the user to configure the parameters of the algorithm. To achieve the rapid convergence of the algorithm, while maintaining the accuracy of the result, it is necessary to use the number of chromosomes in the generation of at least 6 chromosomes and stop the algorithm when changing at least 10 generations.