Exploring regional sustainable commuting patterns based on dockless bike-sharing data and POI data. (June 2022)
- Record Type:
- Journal Article
- Title:
- Exploring regional sustainable commuting patterns based on dockless bike-sharing data and POI data. (June 2022)
- Main Title:
- Exploring regional sustainable commuting patterns based on dockless bike-sharing data and POI data
- Authors:
- Wang, Ruoxuan
Wu, Jianping
Qi, Geqi - Abstract:
- Abstract: The dockless bike-sharing system, which provides a sustainable transportation mode available anytime and anywhere, has played an increasingly important role in urban daily commuting. However, urban structure complexity and job-housing imbalances present difficulties in understanding the characteristics of bicycling travel demand, especially with respect to the commuting patterns in megacities. Moreover, the functional properties of a region implying the different travel purposes of citizens have rarely been considered and incorporated into pattern extraction and analysis. Based on dockless bike-sharing data and point of interest (POI) data, this paper designed a regional indicator system with 24 features and established a set of random forest models to evaluate feature significance for different commuting purposes. The city of Shanghai was used as a case study, and sustainable commuting patterns were regionally extracted. The regional purpose composition can also be effectively derived with the average optimal accuracy of RMSE = 0.0734. The analysis results can help traffic operators and managers not only discover the regional commuting patterns of bike-sharing mobility but also interpret the patterns more specifically and purposefully. In terms of practical applications, the research could benefit the scientific and reasonable allocation of shared bicycles at the regional level to improve individual travel efficiency and comprehensively increase the attractivenessAbstract: The dockless bike-sharing system, which provides a sustainable transportation mode available anytime and anywhere, has played an increasingly important role in urban daily commuting. However, urban structure complexity and job-housing imbalances present difficulties in understanding the characteristics of bicycling travel demand, especially with respect to the commuting patterns in megacities. Moreover, the functional properties of a region implying the different travel purposes of citizens have rarely been considered and incorporated into pattern extraction and analysis. Based on dockless bike-sharing data and point of interest (POI) data, this paper designed a regional indicator system with 24 features and established a set of random forest models to evaluate feature significance for different commuting purposes. The city of Shanghai was used as a case study, and sustainable commuting patterns were regionally extracted. The regional purpose composition can also be effectively derived with the average optimal accuracy of RMSE = 0.0734. The analysis results can help traffic operators and managers not only discover the regional commuting patterns of bike-sharing mobility but also interpret the patterns more specifically and purposefully. In terms of practical applications, the research could benefit the scientific and reasonable allocation of shared bicycles at the regional level to improve individual travel efficiency and comprehensively increase the attractiveness of such a sustainable travel mode. … (more)
- Is Part Of:
- Journal of transport geography. Volume 102(2022)
- Journal:
- Journal of transport geography
- Issue:
- Volume 102(2022)
- Issue Display:
- Volume 102, Issue 2022 (2022)
- Year:
- 2022
- Volume:
- 102
- Issue:
- 2022
- Issue Sort Value:
- 2022-0102-2022-0000
- Page Start:
- Page End:
- Publication Date:
- 2022-06
- Subjects:
- Dockless bike-sharing system -- Commuting patterns -- Regional indicator -- Point of interest -- Random forest model
Transportation -- Periodicals
Telecommunication -- Periodicals
Transport -- Périodiques
Télécommunications -- Périodiques
Telecommunication
Transportation
Periodicals
388 - Journal URLs:
- http://www.sciencedirect.com/science/journal/09666923 ↗
http://www.elsevier.com/journals ↗ - DOI:
- 10.1016/j.jtrangeo.2022.103395 ↗
- Languages:
- English
- ISSNs:
- 0966-6923
- Deposit Type:
- Legaldeposit
- View Content:
- Available online (eLD content is only available in our Reading Rooms) ↗
- Physical Locations:
- British Library DSC - 5069.950000
British Library DSC - BLDSS-3PM
British Library HMNTS - ELD Digital store - Ingest File:
- 22275.xml