Multi-Objective Optimization of Integrated Freight and Passenger Transportation in Shared Autonomous Vehicle Systems

Abstract Recent urbanization and growing e-commerce have ignited freight demand, resulting in transportation challenges such as traffic congestion. A promising solution to the growing freight demand is integrating freight and passenger transportation to reduce the required number of vehicles. Shared Autonomous Vehicle (SAV) systems can efficiently integrate freight and passenger flows by using optimized routes and ride-sharing. Not only vehicle-based integration but also freight-passenger integration in urban spaces such as shared delivery locations (SDLs) such as lockers, would further enhance the performance of the integrated transportation system. The difference in time value between freight and passengers requires us to operate and design integrated transportation systems while explicitly evaluating trade-off relations between passenger convenience and social costs. This paper proposes a multi-objective optimization problem for integrated transportation in SAV systems that captures the dynamic features such as endogenous congestion and ride-share matching of freight and passengers. The optimization model is formulated as a linear programming, allowing us to easily solve it and mathematically derive useful properties for strategic planning. Our numerical experiments with New York City taxi data reveal that the employment of integrated transportation and SDLs synergistically improve passenger convenience and social costs simultaneously.