An Incentive Parking Model For Bike Sharing Users

The bike sharing system renders services to users by allowing users to use bikes for short distance trips.Nevertheless,the bike sharing system faces many challenges,e.g.,the imbalance of the bike distribution among the stations.Some stations are flooded with a lot of bikes while at the same time,some stations do not have enough bikes to satisfy the bike users’ rental demands.Here we mainly focus on the bike sharing system,e.g.,the dock-based and the dockless systems.An Incentive Parking Model(IPM)for users is proposed to ensure the proper parking in the campus of Southeast University,Jiulonghu Campus.This model has three embedded functionalities,i.e.,Load balancing,Payoff function,and Decision model,to ensure the proper management of bike sharing systems in the public.1)Considering the fact that bike sharing systems encounter imbalance problems,the load balance model,which adopts the concept of K-means clustering method to create a list of stations closer to the user’s destination based on the search region,has been designed to solve this problem.The load balance model considers several factors,including the bike parking monetary incentives for bike users allocated among all the stations within the search region,the maximum and minimum station capacities and the capacity of each sub-station.With the load balance model,the user can receive more incentive when his rented bike is returned and parked at the station with fewer bikes and receive less incentive when he parks it at the station with more bikes.The load balance model works together with the Payoff function to uniformly distribute the bikes across all the sub-stations.2)In order to encourage users to follow the guidance of the load balance model,a payoff function is designed to incentivize users.Before a user can receive an incentive,the payoff function makes a comparison to check if there are vacant parking spaces or no vacant spaces for users to park their bikes.Further,the user will not be rewarded if the bike is not parked at the station with no parking space.3)The Decision model is designed to simulate the user’s decisions and movements.Users can obtain a list of stations and the corresponding parameters,e.g.,the walking distance,riding distance,station capacity,and the incentive for each sub-station.However,we encourage users to choose stations with fewer bikes and receive more incentive.The experimental results show that our method can effectively achieve load balancing by motivating bicycle users.