Investigating the two-stage assembly flow shop scheduling problem with uncertain assembling times

The majority of scheduling research considers a deterministic environment with pre- known and fixed data. However, under the tools conditions and worker skill levels in assembly work stations, there is uncertainty in the assembling times of the products. This study aims to address a two-stage assembly flow shop scheduling problem with uncertain assembling times of the products which is assumed to follow a normal distribution. The problem is formulated as an MIP model in general form and under deterministic condition. Since the problem is strongly NP-hard, genetic algorithm is adopted with a new solution structure and fitness function to solve the problem on the practical scales. The presented robust procedure aims to maximize the probability of ensuring that makespan will not exceed the expected completion time. In addition, Johnson’s rule is extended and simulated annealing algorithm is tuned for the problem at hand. The computational results indicate that the obtained robust schedules hedge effectively against uncertain assembling times. The results also show that the proposed genetic algorithm gets better robust schedules than Johnson’s rule and outperforms simulated annealing algorithm in terms of deviation percentage (%D) of the expected makespan from the optimal schedule.