The relocation of carsharing vehicles is one of the main challenges facing its economic viability, in addition to the operational costs and infrastructure deployment. In this paper, we take advantage of an innovative technological proposal of a one-way carsharing system, to test and validate a user-based relocation strategy. The new technology allows vehicles to be driven in a road train by either an operator (up until eight vehicles) or a customer (up to two). The proposed strategy encourages a customer to take a second vehicle along the way, when he/she happens to be moving from a station with excess of vehicles, to a deficient station. As a case study, we have considered a suburban area of the city of Lyon, of which we have a 2015 household travel survey to build a synthetic population undertaking various activities during a day. Then, we inject this population in a detailed multi-agent and multi-modal transport simulation model, to compare the relocation performance of a lower/upper-bound availability algorithm with three other naively intuitive algorithms. The study shows that: (i) relocation algorithm is very sensitive to the ratio of parking slots to fleet size, and (ii) with the right infrastructure we can relocate one vehicle and generate at least one additional trip.
User-Based Relocation of Stackable Car Sharing
Boldrini C;Bruno R;
2019
Abstract
The relocation of carsharing vehicles is one of the main challenges facing its economic viability, in addition to the operational costs and infrastructure deployment. In this paper, we take advantage of an innovative technological proposal of a one-way carsharing system, to test and validate a user-based relocation strategy. The new technology allows vehicles to be driven in a road train by either an operator (up until eight vehicles) or a customer (up to two). The proposed strategy encourages a customer to take a second vehicle along the way, when he/she happens to be moving from a station with excess of vehicles, to a deficient station. As a case study, we have considered a suburban area of the city of Lyon, of which we have a 2015 household travel survey to build a synthetic population undertaking various activities during a day. Then, we inject this population in a detailed multi-agent and multi-modal transport simulation model, to compare the relocation performance of a lower/upper-bound availability algorithm with three other naively intuitive algorithms. The study shows that: (i) relocation algorithm is very sensitive to the ratio of parking slots to fleet size, and (ii) with the right infrastructure we can relocate one vehicle and generate at least one additional trip.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.