An Optimization Model for the Demand-Responsive Transit With Non-Fixed Stops and Multi-Vehicle Type

As a new type of public transportation, demand responsive transit has gradually attracted attention for its flexibility and efficiency. In order to solve problems such as single-vehicle type and fixed stop, and improve its operation efficiency, a collaborative scheduling method combining multi-occupancy vehicle type with non-fixed stop is proposed. Different from the previous studies on scheduling problems of demand responsive transit, which only focused on stop modes such as fixed or non-fixed stop or vehicle types containing single-occupancy or multi-occupancy, this paper also studies the vehicle scheduling of demand responsive transit from the perspective of combination of non-fixed stop and multi-vehicle type. In addition, carbon emission cost is innovatively added into the scheduling model, and an improved genetic algorithm with multiple crossovers within individuals is designed to accelerate the convergence speed of the algorithm and improve the solution efficiency. Finally, taking Shijiazhuang downtown regional road network as an example, the validity of the proposed scheduling method is verified. The results show that compared with the single-occupancy vehicle scheduling method, the operating costs of the multi-occupancy vehicle scheduling method can be reduced by up to 25.0%, and the average passenger in-vehicle time is decreased by up to 8.8%, which could significantly reduce the system operating costs on the premise of ensuring shorter total passenger travel time. Compared with the mode with fixed stops, the average full load ratio of the mode with non-fixed stops increased by 21.7%. Besides, the convergence speed and solving speed of the proposed improved genetic algorithm are increased by 31.7% and 4.8%, compared with the traditional genetic algorithm.