Understanding and Analyzing the Influential Factors on Relocation of Shared Bikes

TL;DR

This paper presents a network flow model for optimizing bike relocation in shared bike systems, demonstrating its effectiveness with real data and analyzing factors affecting profitability and operational decisions.

Contribution

It introduces a novel network flow model for bike relocation and provides a framework for operational and pricing decisions in bikesharing services.

Findings

Relocation cost, bike number, and pricing significantly impact system profitability.

The model effectively optimizes bike redistribution using real-world data.

Sensitivity analysis reveals key factors influencing bikesharing efficiency.

Abstract

To enhance the service quality of bikesharing programs, bike fleet relocation is widely applied to redistribute bikes from bike sufficient areas to bike shortage areas thereby making a better bike-rider balance across different areas. In this study, a network flow model is proposed to solve the optimal relocation problem of shared bikes, and is implemented with the actual dockless shared bike usage data from Yishun, Singapore, to demonstrate its effectiveness. A series of sensitivity analyses are performed to test the impact of the relocation cost, the number of bikes and truck trikes, and the usage price on bike relocation. The results reveal an apparent connection between the profitability of the system and the analyzed factors. This work offers a modeling framework to start and operate a bikesharing service by determining the number of bikes and trikes as well as price schemes. Some bikesharing regulation policies are also suggested.