Load Balancing for Mobility-on-Demand Systems

摘要: In this paper we develop methods for maximizing the throughput of a mobility-on-demand urban transportation system. We consider finite group shared vehicles, located at set stations. Users arrive stations, pick-up and drive (or are driven) to their destination station where they drop-off vehicle. When some origins destinations more popular than others, system will inevitably become out balance: Vehicles build up depleted others. propose robotic solution rebalancing problem that involves empty vehicles autonomously driving between policy minimizes number performing trips. To do this, utilize fluid model customers in The takes form nonlinear time-delay differential equations. then show optimal can be found as linear program. By analyzing dynamical model, every reaches an equilibrium which there excess no waiting customers.We use real-time operate highly variable environments. verify performance simulated environment with stochastic features real-world networks.