Enhancing On-demand Transportation: Prioritizing Travel Costs with Hybrid Evolutionary Algorithms

Abstract

Objective: Enhancing sustainability in on-demand transportation is a challenging task by offering flexible and optimized mobility solutions. This study focuses on on-demand mobility systems. We introduce a novel evolutionary computing method designed to address a bi-objective customized on-demand transportation problem, aiming to minimize total travel cost and total waiting time. Methods: The method incorporates specific optimization techniques, along with an efficient dominance sorting approach, using an intelligent candidate list to reduce computational time. Results: Comparative results demonstrate the effectiveness of this hybrid computing method, particularly when prioritizing total travel cost from the service provider’s perspective. Conclusion: These findings present a promising framework for decision-makers, empowering them to navigate favorable compromises between conflicting objectives and make informed choices aligned with their preferences and customer-oriented constraints.