Atac, S., Obrenovic, N., and Bierlaire, M. (2022)

Evaluating different strategies to solve rebalancing operations in car sharing systems

22nd Swiss Transport Research Conference, Ascona, Switzerland

Car sharing (CS) services have become popular due to their financial and environmental benefits. The CS operators have offered flexibility by allowing one-way trips which resulted in vehicle imbalance in the service area. They have then introduced rebalancing operations to reduce the imbalance, and thus, to increase the level of service. The methods studied in the literature focus on forecasting the demand to determine the rebalancing strategy. This work proposes a framework which compares different strategies to solve rebalancing operations in one-way station-based car sharing systems in terms of cost and level of service. One of the crucial components of this framework is a demand model that represents the daily flow in the network. Instead of collecting the trip demand data, we feed the trip demand output of Multi-Agent Transport Simulation Toolkit (MATSim) as an input to our framework. This also allows us to explore the different uncertainties that can occur in the system, such as fluctuations in trip demand. The results of the framework help the decision maker to better analyze the system and choose the best rebalancing strategy under different scenarios.This paper focuses on bike sharing systems with static rebalancing operations. Two multi-objective mathematical models are specifically crafted for the rebalancing-oriented clustering problem. These models and two agglomerative hierarchical clustering approaches are compared with respect to resulting cost of rebalancing operations.