Effect of built environment on shared bicycle reallocation: A case study on Nanjing, China. (October 2019)

• Record Type: Journal Article
• Authors: Zhao, De; Ong, Ghim Ping; Wang, Wei; Hu, Xiao Jian
• Is Part Of: Transportation research. Volume 128(2019)
• Page Start: 73
• Page End: 88
• Publication Date: 2019-10
• Subjects: Built environment -- Bicycle sharing -- Bicycle demand -- Reallocation -- Zero-inflated negative binomial regression; Transportation -- Research -- Periodicals; 388.011
• DOI: 10.1016/j.tra.2019.07.018
• Languages: English
• ISSNs: 0965-8564

Abstract

Abstract: Shared bicycles provide a convenient mobility option to commuters especially for short-distance trips. Nevertheless, it also presents a challenge to bicycle-sharing operators as they have to deal with reallocation issues, i.e. removing bicycles from parking facilities which are at or near capacity and refilling parking facilities that are in need of bicycles. Few studies in the literature have actually tried understanding why certain docking stations are prone to excessive demand or suffer from a lack of parking supply. This paper attempts to identify the demographic, built-environment and transport-infrastructure indicators that can potentially aid policy-makers and operators in identifying parking facilities prone to bicycle reallocation. In particular, we have adopted the bicycle sharing operations in Nanjing, China as a case study to understand how such indicators can be identified for appropriate parking infrastructural enhancements. To achieve this goal, this study has established zero-inflated negative binomial models using multi-source data including point-of-interest (POI), daily weather, transit stop location, demographic data and bike-share smart card data. The model results obtained from this study suggest that built environment correlates significantly to shared bicycle reallocation count. In general, bicycle docking stations with large reallocation counts are more likely to be found near residences, bus stops, metro stations, employment areas, restaurants, amenities, parks, sports facilities, and clinics/hospitals; while stations near entertainment facilities, places of attraction, hotels, shopping malls, and educational institution tend to have balanced demand and supply. Analysis on the elasticity values revealed that mean temperature and station capacity are the most influential factors in bicycle reallocation. Among all POIs, presence of restaurants and areas with high employment tend to exhibit strongly a need for morning bicycle removal and evening bicycle refilling at docked stations. Policy makers can provide actual guidelines in the planning of shared bicycle parking facilities using the findings and methodologies presented in this study.