This paper proposes a new framework to tackle computer-aided modeling and evolutionary algorithms into conventional design phase of a machine, which, in turn, significantly reduces time and cost of structural optimization. The model-based engineering approach overcomes the crudity of hard modeling, field experiments and statistical analysis for finding the optimum structure of a design. Genetic algorithm incorporated with fuzzy modeling established a hybrid computational algorithm which predicts optimal sizing of a platform, developed for a chickpea harvester header. The harvesting losses of the platform’s configurations in field trials were fed to the metaheuristic approach to develop a soft simulator for redesigning of the machine. Acceptable harvesting performance of the optimized harvester in field trials confirmed the robustness feature of the experiments based simulator. Further the results validated the virtual model and verified the reliability of the automatically generated harvester. The methodology can be employed for structural optimization of mechanical systems.
