Dynamic Scheduling of Electric Demand Responsive Transit Considering Time‐of‐Use Pricing

Demand‐responsive transit (DRT) has been widely adopted in many cities worldwide due to its advantages of flexible routing and personalized services. However, with the progressive replacement of fuel‐powered buses by electric buses, how to synergistically optimize vehicle charging schedules with DRT dynamic dispatching has emerged as a critical challenge. To address the issue, this study proposes a dynamic dispatching approach for electric DRT considering time‐of‐use (TOU) pricing. Specifically, integrated with a rolling horizon optimization framework, a coordinated optimization model for vehicle routing and charging scheduling is established based on TOU electricity pricing, with the objectives of minimizing passenger delay costs, vehicle operating costs, charging costs, and peak‐hour charging penalty costs. The problem is decomposed into a static phase and a dynamic phase, for which an adaptive large neighborhood search (ALNS) algorithm and a heuristic insertion algorithm are respectively designed to achieve efficient solutions. Case validation based on taxi order data from Shenzhen, China demonstrates that compared with the hard time window model, the proposed soft time window model reduces the total cost by 11.3% and the passenger delay cost by 18.2%, while improving the service rate from 54.0% to 64.9%. Compared with the traditional reactive charging strategy, the proposed model achieves a 20.7% reduction in total cost, a 23.0% decrease in passenger delay cost, and a 21.4% drop in charging cost. In contrast to the alternative model that excludes charging costs from the optimization objective, the proposed model reduces the total cost by 8.0%, the charging cost by 14.7%, and the peak‐hour charging penalty cost by 25.6%, demonstrating a distinct capability to guide vehicles toward charging during off‐peak electricity periods and significantly improving the utilization of low‐price electricity. These findings provide practical insights for the actual operational dispatching of electric DRT systems.