WIT Press

Utilising dedicated pedestrian networks to understand the relationship between accessibility and pedestrian density in Asian cities


Free (open access)


Volume 5 (2021), Issue 2



Page Range

150 - 161

Paper DOI



WIT Press


Daniel M. Pearce, Ryoji Matsunaka, Tetsuharu Oba & Shota Yamamoto


As pedestrian networks tie together all transportation modes, their accessibility is crucial to supporting public transportation and fostering more sustainable and liveable cities. Nevertheless, most studies have relied on street networks when calculating pedestrian accessibility. While street networks can substitute for pedestrian networks in some cases, they fail to capture all paths available to pedestrians. Recent studies argue that this may distort our understanding of connectivity, accessibility and consequently the walkability of an area. This study sought to understand the effect of pedestrian network accessibility on segment-scale pedestrian density. To achieve this goal, dedicated pedestrian networks were constructed for two 400 m2  areas in Bangkok and Osaka. In each site, the effect of accessibility variables, namely reach and gravity, were assessed first before controlling for other environmental factors addressing con- nectivity, transit proximity and path conditions. Measures were evaluated using ArcGIS and the Urban Network Analysis toolbox using data obtained from OpenStreetMap databases and field surveys. The results show that accessibility variables alone can account for up to 33% and 35% of observed pedes- trian densities in Bangkok and Osaka, respectively. Accessibility to retail land uses is the biggest driver in this relationship. However, when controlling for all variables, only in the denser, more connected Osaka site did accessibility to retail remain significant, helping to explaining up to 45% of observed pedestrian densities, while in Bangkok, its effect was minimal with proximity to rail transit being the only significant variable.


built environment, connectivity, dedicated pedestrian networks, OpenStreetMap, pedestrian accessibility, pedestrian density, Urban Network Analysis