Research On The Theory And Case Study Of Redistribution Operation Model Of Bicycle-sharing System (BSS)

Since Amsterdam first launched a bicycle-sharing system (BSS) in1965, a numberof cities have established BSS. Until the BSS ‘Vélib’ in Paris was successfully operatedin2007, BSS began to really get more and more attention of society and researchers.The stations of BSS are usually located in bicycle-trip-focused sites, andbicyclesare placed at BSS stations for people to use, and at each station there is a smart terminalto manage and record people’s borrowing and returning behavior. Bicycles can be usedby applying specified cards to gain the right to use public bicycles.As the tide of usagedemand, one-way trips within a certain period will result in bicycles disproportionallydistributed, and the imbalanced problem can reduce the level of service of BSS. In orderto alleviate the unbalanced problem of bicycles, many scholars studied it in two aspects:station planning and operation, and they have achieved good results.This paper focuses on bicycle imbalance among stations, and discusses the stationdesign and operation distribution, and it has in-depth analysis in bicycle redistributionproblem. A redistribution scheduling optimization model is developed and a case isanalyzed to have deeper study. The model is coded in MATLAB, and applies improvedant colony algorithm and genetic algorithm to find the optimal solution.The main resultsof this study include:1. Introduce the method of station planning based on the demand of borrowing andreturning demand at stations, supposing that the location of stations has beendetermined. In addition, it develops principles to calculate feasible region ofredistribution amount.2. It develops a redistribution routing model. This model describes theredistribution amount when visiting a station, the conditions of redistribution tasksaccomplished and the changes of the number of bicycles in the truck. The objective ofthe model is to minimize the travel length of redistribution while the truck does its bestto redistribute bicycles among stations, and it searches the optimal solution through animproved ant colony algorithm.3. It develops a redistribution scheduling optimization model. The model considersthe interaction of redistribution amount, partition of stations and redistribution path, and it is developed based on station planning and redistribution routing model. Theobjective of the comprehensive optimization model is to achieve the minimumgeneralized operation cost.4. BSS in Wuhan is studied. In the case study, the redistribution scheduling isoptimized and the paper gives quantitative analysis about the signification of partition inlarge-scale redistribution, mainly in the improvement of required time to completeredistribution (redistribution efficiency) and performance of BSS.Finally, it summarizes the full paper, and discusses future research directions aboutredistribution operational optimization model.