Deep reinforcement learning of passenger behavior in multimodal journey planning with proportional fairness

Abstract Multimodal transportation systems require an effective journey planner to allocate multiple passengers transport operators. One example is mobility-as-a-service, a new mobility service that integrates various modes through single platform. In such multimodal and diverse planning problem, accommodating heterogeneous with different dynamic preferences can be challenging. Furthermore, may behave based on experiences expectations, in the sense experience affects their state decision of next service. Current methods treating each optimization as non-time varying problem cannot adequately model passenger memories over many journeys time. this paper, we Markov where prior have transient effect future long-term satisfaction retention rate. As such, formulate multi-objective considers individual preferences, experiences, memories. The proposed approach dynamically determines utility weights obtain optimal plan for status. To balance profit received by operator, present variant-based proportional fairness. Our experiments using real-world synthetic datasets show our enhances satisfaction, compared baseline methods. We demonstrate overall increased 2.3 times, resulting higher rate caused levels. facilitate participation operators promote acceptance MaaS.