Data-Adaptive Simulation: Cooperativeness of Users in Bike-Sharing Systems

Bike-sharing systems undergo a rapid expansion due to technical improvements in the operation combined with an increased environmental and health awareness of people. The acceptance of such system depends heavily on the availability of bikes at stations. In spite of truck-based redistribution efforts by the operators, stations still tend to become empty/full, especially in rush-hour situations. In this paper, we explore an incentive scheme that encourages users to approach nearby stations for renting or returning bikes, thereby redistributing them in a self-organized fashion. A cooperativeness parameter is determined by the fraction of users that responds to an incentive by choosing the proposed stations. The microscopic simulations of the actual bike-sharing system is based on data taken from Washington, D.C. (2014). From these data, stochastic parameters can be determined such as the rush of users for a station given as a function over time. Here, we propose a data-adaptive simulation approach to measure the impact of different cooperativeness parameters. The proposed approach realizes a data-adaptive simulation where the knowledge/data space of the application is filled on demand. If knowledge/data already exist, no further simulations are required. If not, the required microscopic simulations are executed and the data set is enriched with their results.