A hybrid GA-simulation approach to improve JIT systems

This study presents a hybrid approach involving genetic algorithms (GAs) as an optimisation search technique and a simulation model, representing the dynamic behaviour of a system and its limitations, to improve an existing just-in-time (JIT) manufacturing system. To achieve the objective, first, the existing system is modelled and simulated (by considering the system's limitations and its dynamic behaviour). Second, the integrated simulation model is tested and validated by analysis of variance. Third, the hybrid GA-simulation approach is used in an interactive manner to determine the optimal/near-optimal number of kanban cards in different stations of the existing JIT system. The presented hybrid approach is tested and applied to an auto industry production line. Furthermore, it is compared with the practical JIT through analysis of variance (ANOVA) and the results show improvements in the average daily production rate, the average resource utilisation and the average cycle time but some deterioration in the average queue length and in-process inventory is inevitable.