Can bike sharing achieve self-balancing distribution? Evidence from dockless and station-based cases

In many cities, bike sharing systems, including station-based bike sharing (SBBS) and dockless bike sharing (DBS), are gaining popularity rapidly. Bike rebalancing is one of the most expensive aspects of bike sharing operations, and it takes several hours. In terms of reducing the inefficiencies of frequent short-term bike rebalancing, whether bike distribution achieves long-term self-balance for one day or even longer periods is a critical issue that has received insufficient attention. This paper aims to provide insights into long-term facility planning by investigating the self-balancing phenomenon of shared bikes. It is evaluated using daily stability analyses from the DBS case in Nanjing, China, and the SBBS case in New York, USA. DBS virtual stations were identified throughout the city, and (virtual) stations can be classified into four categories using various clustering methods. The findings demonstrate that 72% of DBS virtual stations and 81% of SBBS stations can achieve bike selfbalancing, with only a few (virtual) stations failing to do so. In terms of non-self-balancing stations, bikeincreasing stations are primarily located in the city center, whereas bike-fluctuating stations are primarily found near metro lines. This research can assist bike sharing companies with their daily operations and contribute to government management.