A study of the robustness of the group scheduling method using an emulation of a complex FMS

In the field of predictive-reactive scheduling methods, group sequencing is reputed to be robust (in terms of uncertainties absorption) due to the flexibility it adds with regard to the sequence of operations. However, this assumption has been established on experiments made on simple theoretical examples. The aim of this paper is to carry out experimentation on a complex flexible manufacturing system in order to determine whether or not the flexibility of the group scheduling method can in fact absorb uncertainties. In the study, transportation times of parts between machines are considered as uncertain. Simulation studies have been designed in order to evaluate the relationship between flexibility and the ability to absorb uncertainties. Comparisons are made between schedules generated using the group sequencing method with different flexibility levels and a schedule with no flexibility. This last schedule takes into account uncertainties whereas schedules generated using the group sequencing method do not. As it is the best possible schedule, it provides a lower bound and enables to calculate the degradation of performance of calculated schedules. The results show that group sequencing perform very well, enabling the quality of the schedule to be improved, especially when the level of uncertainty of the problem increases. The results also show that flexibility is the key factor for robustness. The rise in the level of flexibility increases the robustness of the schedule towards the uncertainties.