An Artificial-Neural-Network-Based Model for Real-Time Dispatching of Electric Autonomous Taxis

This paper presents a real-time dispatching model for electric autonomous vehicle (EAV) taxis that combines mathematical programming and machine learning. The EAV taxi problem is formulated solved as an integer linear program maximizes the total reward serving customers. optimal dispatch solutions are generated by simulating dispatched optimization model. artificial-neural-network-(ANN)-based was trained using model’s to learn strategies. Although decisions made ANN-based not optimal, system’s performance very close in terms of customer service taxis’ operational efficiency. In addition, runs much faster. By comparing with current taxis, it found our can improve efficiency reducing empty travel distance. also reduce fleet size 15% while maintaining comparable level fleet.