Optimizing stochastic multi-project scheduling with a simulation integrated multi-objective genetic algorithm

The significance of project scheduling and sequencing has increased considerably in recent years, driven by the rising customer demand for highly personalized solutions. Companies have to consider multiple criteria while executing multiple projects simultaneously to meet the customer demands. Therefore, this study focuses on the multi-objective multi-project scheduling and sequencing problem (MP-SSP). A simulation-based mathematical model is developed and integrated with a multi-objective genetic algorithm. The objectives of this model are to minimize the project execution cost, project completion time and project lateness simultaneously while maximizing the resource utilization in the stochastic environment. Goal attainment programming is introduced in the simulation integrated multi-objective genetic algorithm (SIHMO-GA) to increase the effectiveness of the algorithm. Further, response surface methodology (RSM) has been used to find the optimum parameters of the proposed SIHMO-GA. The effectiveness of the proposed SIHMO-GA is evaluated through a real-world case study by comparing it with simulation-optimization approaches, namely the multi-objective genetic algorithm (MOGA) and goal attainment programming. Gap analysis indicates that the SIHMO-GA provides best trade off values of the above-mentioned conflicting objectives under a stochastic environment. This study supports practical scheduling and sequencing of multiple projects in a stochastic environment by generating solutions that maximize profit, enhance resource utilization, and ensure customer satisfaction through timely project delivery.