2018/10/18 by Lina Al‐Kanj, Juliana Nascimento, Al-Kanj, Lina +3 · 4 citations
Engineering · #Electric Vehicles and Infrastructure #Transportation and Mobility Innovations #Smart Grid Energy Management
paper · pdf · doi:10.48550/arxiv.1810.08124
Within a decade, almost every major auto company, along with fleet operators\nsuch as Uber, have announced plans to put autonomous vehicles on the road. At\nthe same time, electric vehicles are quickly emerging as a next-generation\ntechnology that is cost effective, in addition to offering the benefits of\nreducing the carbon footprint. The combination of a centrally managed fleet of\ndriverless vehicles, along with the operating characteristics of electric\nvehicles, is creating a transformative new technology that offers significant\ncost savings with high service levels. This problem involves a dispatch problem\nfor assigning riders to cars, a surge pricing problem for deciding on the price\nper trip and a planning problem for deciding on the fleet size. We use\napproximate dynamic programming to develop high-quality operational dispatch\nstrategies to determine which car is best for a particular trip, when a car\nshould be recharged, and when it should be re-positioned to a different zone\nwhich offers a higher density of trips. We prove that the value functions are\nmonotone in the battery and time dimensions and use hierarchical aggregation to\nget better estimates of the value functions with a small number of\nobservations. Then, surge pricing is discussed using an adaptive learning\napproach to decide on the price for each trip. Finally, we discuss the fleet\nsize problem which depends on the previous two problems.\n