Measuring pedestrian flows in public spaces: Inferring walking for transport and recreation using Wi-Fi probes

• Published in: Building and environment 2023-02, Vol.230, p.109999, Article 109999
• Main Authors: Hou, Jingxuan, Zhang, Enjia, Long, Ying
• Format: Article
• Language: English
• Subjects: Pedestrian flow; Public space; Recreational walking; Transport walking; Walkability; Wi-Fi probes
• ISSN: 0360-1323; 1873-684X
• DOI: 10.1016/j.buildenv.2023.109999

Differentiating transport and recreational walking in public spaces could promote the precise design of walkable public spaces for different walking demands to encourage more walking behaviors. However, previous studies mainly relied on field observations and self-reports, failing to quantitatively distinguish and depict the spatial-level usage of transport and recreational walking of all pedestrians passing by. This study proposed an approach based on the traffic counts and the number of pedestrians to infer transport and recreation walking in public spaces. A comparative experiment using Wi-Fi probes to collect pedestrian data in a gated residential community and a creative center in Beijing, China, was conducted to verify the applicability of this method. The results demonstrated that the transport walking index (the number of pedestrians) could portray the volume of transport walking, and the recreational walking index (average traffic counts of each pedestrian) could depict the proportion of recreational walking in public spaces. Two tests using different time threshold parameters and field observations verified the robustness of the results. Given the low-cost and long-duration observation, this method can potentially support the process of Post Occupancy Evaluation and Environment and Behavior research in more public spaces to make them more walkable. •A novel method was proposed to quantitatively differentiate walking for transport and recreation in public spaces.•A comparative study was conducted to verify the applicability of the method.•Tests using different parameters and field observations verified the robustness of the results.•Residential communities might have more complicated and mixed behaviors.