Assembly Operation Optimization Based on A Hybrid Particle Swarm Optimization and Genetic Algorithm

The assembly dimensional quality can be improved by optimizing the assembly operations between parts. Firstly, as there is large quantity of geometric feasible assembly sequences for the auto-body, the multi-attribute directed liaison graph is applied to describe the precedence relationships and assembly control characteristics between parts. Secondly, with assembly deviation propagation as the fitness function, the hybrid particle swarm optimization and genetic algorithm is proposed to optimize the assembly operations between parts. Finally, the optimal assembly sequence is selected through assembly variation propagating based on linear assembly variation analysis model. The optimization of the key control characteristics is illustrated by the auto-body side assembly.

---

W artykule zaproponowano wykorzystanie hybrydowego połączenia optymalizacji rojem cząstek i algorytmu genetycznego w optymalizacji procesu montażu/łączenia części. Metoda bazuje na funkcji użyteczności wynikającej z propagacji błędu złożenia. Efekty działania zaprezentowano na przykładzie składania boku karoserii samochodu.