116. Decision Tree Model In Dynamic Vehicle Dispatching in Demand Resp…

International Oral Session, International, 27(금) 13:20-15:00, 라카이볼룸Ⅰ 

Decision Tree Model In Dynamic Vehicle Dispatching in Demand Responsive Transit Considering Vehicle Rebalancing / Tandan Sanjay(Korea National University Of Transportation), Seunghak Kim(Korea National University Of Transportation), Morris Alain Anthony(Korea National University Of Transportation), Hyun Kim(Korea National University Of Transportation)

This paper presents a Decision Tree Model designed to optimize the efficiency of Demand-Responsive Transit (DRT) systems through dynamic vehicle dispatching and strategic re-balancing. DRT systems offer flexible and adaptive transportation services by responding to real-time passenger demands, making them particularly effective in low-density and off-peak areas. The proposed model utilizes historical call data and employs a Random Forest approach to predict spatio-temporal demand patterns. This allows for proactive repositioning of idle vehicles, reducing passenger wait times and maximizing fleet utilization. By integrating real-time decision-making with advanced predictive analytics, the model significantly improves the reliability and user experience of DRT services. The findings suggest that this approach enhances service quality and supports the broader goals.