Modeling workers rotation in divisional seru production systems

This study proposes a mathematical model for the job rotation problem in the Divisional seru Production System (DSPS) and develops an efficient solution algorithm. DSPS, a transitional phase toward a fully realized seru system, enhances flexibility and workforce adaptability in volatile manufacturing environments. A non-linear programming model optimizes maximum flow time in job rotation scheduling. Small-scale instances are solved using GAMS, while the Invasive Weed Optimization (IWO) meta-heuristic handles medium- and large-scale cases. Results show that IWO significantly outperforms GAMS in computation time while maintaining solution accuracy, with differences in objective values under 5% for most cases. Additionally, in 62.5% of cases, the number of assigned workers is fewer than the initial number of workers provided for each problem.

Randomly generated test instances validate the model and algorithm, confirming their effectiveness in reducing flow time and workforce requirements. Post-optimal trials indicate that the number of rotation periods can be adjusted to minimize the flow time. It was discussed that when the number of rotation periods is optimized to minimize the flow time, imbalances among cells are also minimized. This study fills a gap in the literature and provides new insights for optimizing DSPS.