Processing Time Optimization for Robot Applications

Abstract

This research work includes the use of heuristic algorithms to automatically generate processing time optimized robot programs for manufacturing processes in the automotive industry. For this, we've implemented a genetic algorithm with multi-parent recombination and adjacency-based crossover. A reallocation mutation is also introduced to optimize the load balancing by classifying tasks into common and fixed tasks depending on their location relative to the robots' workspace. The heuristic is compared to an exact solver by applying it to a test problem. Lastly, the methodology is also applied to a real business problem in the area of vehicle sealing.