Optimizing a Dynamic Transport on Demand Problem Using the Multi-Objective Particle Swarm Optimization Algorithm

This paper introduces a Multi-objective Particle Swarm Optimization (MOPSO) algorithm to address the Transport on Demand Problem (TDP). The dynamic nature of the TDP, which includes challenges such as vehicle breakdowns and traffic congestion, complicates the optimization of transportation systems. Our approach focuses on efficiently assigning passengers to vehicles while minimizing travel distance and ride time. We compare the performance of the proposed MOPSO algorithm with established multi-objective optimization algorithms, including NSGA-III, NSGA-II, and SPEA2, using a variety of Pareto front metrics , such as Hypervolume, Spacing, and R2 Indicator. The results show that MOPSO outperforms these algorithms, offering superior solutions in terms of both efficiency and solution quality. This research contributes to the development of optimized transport systems that improve the efficiency and quality of on-demand services, helping to address the growing demand for more responsive and effective transportation solutions in modern cities.