详情

原文传递 Different Model Considerations for Repositioning Free--Float Bike Sharing Systems with Faulty Bikes

论文题名：	Different Model Considerations for Repositioning Free--Float Bike Sharing Systems with Faulty Bikes
关键词：	Bike-sharing system;Repositioning problem;Simulated annealing algorithm
摘要：	The Free-Floating Bike-sharing System(FFBS),with the advantages of highly flexible mobility,less usage of fuel,and the reduction of congestion and emission,is considered as an essential component of future green and multimodal transportation.It assists commuters to pick up and drop off a rented bike without going to a docking station before and after the ride.However,the convenience of FFBS can also be a problem for users and operators because the randomly scattered usable and faulty bikes can bring about bikes’unavailability and maintenance difficulties.This imbalance against user demand and widespread faulty bikes not only result in a negative impact on bike-sharing companies’service quality but also cause users’safety concerns.In this study,a shared bike repositioning framework is presented to collect faulty bikes to the depot,and to distribute repaired bikes among different stations to obtain optimal inventory level at all nodes by repositioning vehicles.This framework is realized by minimizing the total routing cost using optimum fleet size supported by various time and capacity constraints. 　　The presented formulation is an extension to the traveling salesman problem(TSP)and proved to be strictly NP-hard.So,the computation time increases exponentially with the increase in the number of stations.Hence,to solve the problems in a reasonable time,a simulated annealing algorithm is developed in this thesis.The presented algorithm gives good results in reasonably less time as compared to the exact solutions through CPLEX solver. 　　Furthermore,a free-float bike-sharing system repositioning problem with faulty bikes using battery electric vehicles(BEVs)is presented.The target inventory of bikes at each station is obtained while minimizing the total traveling time through the presented formulation.Using CPLEX solver,the model is demonstrated through numerical experiments considering the various vehicle and battery capacities,and a cost-benefit analysis is performed for BEV and conventional internal combustion engine vehicles(ICEVs)while taking BEV manufacturing and indirect emission into account.The results show that the annual cost incurred on an ICEV is56.9%more than the cost of using an equivalent BEV.Moreover,the life cycle emissions of an ICEV are48.3%more as compared to an equivalent BEV.Moreover,the operational cost of a BEV reduces significantly with the increase in battery capacity. 　　In addition,the former model is extended for two real-time considerations.First,shifting faulty bikes to the stations determined by the model instead of shifting to their closer stations.Second,considering a flexible target demand interval at each station instead of a fixed target inventory at each station.The results show that the collection of faulty bikes to their nearest stations results in shorter handling distances.However,the total operational cost may increase because of the longer paths of the repositioning vehicle.Similarly,flexibility in the next day target demand allows the vehicle to choose a target inventory level from the given interval which minimizes the total cost.Whereas,a fixed demand limits the vehicle to ensure the fixed inventories at each station which may result in longer traveling time and higher operational cost.
作者：	Muhammad Usama
专业：	Master of Engineering
导师：	Yongjun Shen
授予学位：	硕士
授予学位单位：	东南大学
学位年度：	2019
正文语种：	中文