solving a bi-objective flexible flow shop problem with transporter preventive maintenance planning and limited buffers by nsga-ii and mopso

This study deals with a bi-objective flexible flow shop problem (boffsp) with transportation times and preventive maintenance (pm) on transporters via considering limited buffers. the pm actions on transporters are a missing part of the literature of the flexible flow shop problem (ffsp) which before the breakdown occurs, each transporter at each stage is stopped and the pm action is performed on it. the capacity of each intermediate buffer is limited and each job has to wait in the intermediate buffers. by including all these features in the proposed boffsp, not only processing times affect the objective functions, but also, the transportation times of jobs, the waiting time of jobs in the intermediate buffers, and availability of transporters in the system are considered in the model and make it a sample of a real-world ffsp. the presented boffsp has simultaneously minimized the total completion time and the unavailability of the system. as the problem is np-hard, a non-dominated sorting genetic algorithm ii (nsga-ii) and a multi-objective particle swarm optimization (mopso) is proposed to solve the model for large size problems. the experimental results show that the proposed mopso relatively outperforms the presented nsga-ii in terms of five different metrics considered to compare their performance. afterwards, two one-way anova tests are performed. it can be observed mopso achieves relatively better results than nsga-ii. finally, sensitivity analysis is conducted to investigate the sensitivity of the objective functions to the number of jobs and their transportation time at each stage.