Login / Signup

Who changed and who maintained their urban bike-sharing mobility after the COVID-19 outbreak? A within-subjects study.

Rudi SeifertMaite Pellicer-ChenollLaura Antón-GonzálezMiquel PansJosé Devís-DevísLuis-Millán González
Published in: Cities (London, England) (2023)
The COVID-19 pandemic has remarkably impacted urban mobility. All non-essential movements were restricted in Valencia (Spain) to contain the virus. Thus, the transport usage patterns of Valencia's bike-sharing system (BSS) users changed during this emergency situation. The primary objective of this study was to analyse the behaviour patterns of BSS users in Valencia before and after the COVID-19 outbreak, specifically those who maintained or changed their transport routines. A within-subjects comparison design was developed using a group of BSS users before and after the onset of the pandemic. Data mining techniques were used on a sample of 4355 regular users and 25 variables were calculated to classify users by self-organising maps analysis. The results show a significant reduction (40 %) in BSS movements after the outbreak during the entire post-outbreak year. There was some recovery during the rest of 2020; however, this has yet to reach the pre-pandemic levels, with variations observed based on the activities performed in different areas of the city. Of the users, 63 % changed their BSS use patterns after the onset of the pandemic (LEAVE group), while 37 % maintained their patterns (REMAIN group). The user profile of the REMAIN group was characterised by a general reduction of approximately 35 % of journeys during 2020, with a slight increase in morning movements compared to those made in the evening. These users also presented an equivalent number of cycling days to those of the previous year, reduced the number of connections and increased the network's density and the travelling speed. These results can be useful in estimating the percentage of people who do not vary their usual behaviour during emergencies. Finally, several policy implications are outlined based on the findings.
Keyphrases
  • sars cov
  • coronavirus disease
  • public health
  • healthcare
  • health information
  • social media
  • machine learning
  • mental health
  • deep learning
  • data analysis