Impact of data processing on deriving micro-mobility patterns from vehicle availability data

Summary, in English

Vehicle availability data is emerging as a potential data source for micro-mobility research and applications. However, there is not yet research that systematically evaluates or validates the processing of this emerging mobility data. To fill this gap, we propose a generally applicable data processing framework and validate its related algorithms. The framework exploits micro-mobility vehicle availability data to identify individual trips and derive aggregate patterns by evaluating a range of temporal, spatial, and statistical mobility descriptors. The impact of data processing is systematically and rigorously investigated by applying the proposed framework with a case study dataset from Zurich, Switzerland. Our results demonstrate that the sampling rate used when collecting vehicle availability data has a significant and intricate impact on the derived micro-mobility patterns. This research calls for more attention to investigate various issues with emerging mobility data processing to ensure its validity for transportation research and practices.

• Author: Pengxiang Zhao; He Haitao; Aoyong Li; Ali Mansourian
• Department/s: Dept of Physical Geography and Ecosystem Science; Centre for Geographical Information Systems (GIS Centre)
• Publishing year: 2021-08
• Language: English
• Publication/Series: Transportation Research Part D: Transport and Environment
• Volume: 97
• Links: http://dx.doi.org/10.1016/j.trd.2021.102913
• Document type: Journal article
• Publisher: Elsevier
• Topic: Earth and Related Environmental Sciences; Engineering and Technology
• Keywords: Data processing; Data sampling; E-scooter sharing; GPS; Micro-mobility; Spatio-temporal patterns; Trip identification; Vehicle availability data
• Status: Published
• ISBN/ISSN/Other: ISSN: 1361-9209