Scholarlink

Service quality is strongly related to user satisfaction and is an essential factor for service providers to develop marketing strategies. In the case of mobility services, a number of new services have appeared in the last decade as a result of the Mobility as a Service (MaaS) trend, and new marketing strategies are required in order to survive against the competition and acquire users. In this study, we conduct a multi-agent simulation for mobility services to reveal how service quality influences user acquisition. The model of the simulation contains two types of agents, service providers and users, which are subjected to reinforcement learning toward each other. The simulation results demonstrated the following: (1) Users tend to prefer services with high availability. (2) Less popular services can acquire a certain number of users by differentiating themselves from the dominant service. (3) Services can attract and retain a large number of users by running campaigns during their launch.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Multi-agent Reinforcement Learning for Market Competition in Mobility Services

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Haruki Sato,
  • Fujio Toriumi,
  • Yukihisa Fujita

摘要

Service quality is strongly related to user satisfaction and is an essential factor for service providers to develop marketing strategies. In the case of mobility services, a number of new services have appeared in the last decade as a result of the Mobility as a Service (MaaS) trend, and new marketing strategies are required in order to survive against the competition and acquire users. In this study, we conduct a multi-agent simulation for mobility services to reveal how service quality influences user acquisition. The model of the simulation contains two types of agents, service providers and users, which are subjected to reinforcement learning toward each other. The simulation results demonstrated the following: (1) Users tend to prefer services with high availability. (2) Less popular services can acquire a certain number of users by differentiating themselves from the dominant service. (3) Services can attract and retain a large number of users by running campaigns during their launch.