A New Approach of Integrating Evolutionary Computation Method Decision Logic into a Conventional Simulation Environment

Abstract The present manuscript investigates the research problem of the integration of Evolutionary Computation method optimization results into a conventional simulation environment, with the main purpose of overcoming the conventional decision logic of the simulation environment. Simulation model interactivity was described as a research problem relating to Evolutionary Computation. Ten Brandimarte and five Kacem benchmarks were applied for calculation of optimization results suitable for transfer to the simulation model. The C-metric comparison method was used to evaluate Multi-Objective Heuristic Kalman Algorithm (MOHKA) Pareto solutions’ efficiency in relation to the Multi-Objective Particle Swarm Optimization (MOPSO) and Bare-Bones Multi-Objective Particle Swarm Optimization (BBMOPSO) algorithms. A new simulation modeling interactive block diagram is presented, along with a description of its characteristics, advantages and limitations. The obtained optimization results of the MOHKA algorithm show high capability to solve complex multi-objective optimization problems, especially on real-world evaluated manufacturing systems’ data. The obtained optimization and simulation results show a new robust method of optimization data transfer between a mathematical optimization algorithm and a conventional simulation modeling environment. The main contributions are related to an original research question, showing that the presented approach can overcome the integrated decision logic of conventional simulation software, and transfer the optimization results of the Evolutionary Computation method into the simulation model. The presented integrating approach can be used in a variety of production systems when the need arises of transferring optimization results into a simulation model.