A Data-Driven Study on Pedestrian Walking Behaviour as Transitioning Different Spaces

Pedestrians' daily walking behaviours involve transitioning urban spaces with different functions in the pedestrian transportation. Current research has extensively investigated different pedestrian walking behaviours. However, beyond the influence of physical disturbances and structure on pedestrian behaviour, current research lacks an understanding of the impact and role of space itself on pedestrian behaviour from the perspective of spatial changes, and similarities in the behaviour of pedestrians as they transition different spaces. This study uses real-world datasets to conduct a data-driven study on the impact of spaces (open space, sidewalk, and crossing) on pedestrian walking speed. The results show that, pedestrian speed changes statistically significantly near the connection line between two spaces. The location of this significant change in speed depends on the type of connected space. Also, the magnitude of speed change in this area depends on the type of adjacent space and the movement direction. We also study the extent to which the turning behaviour of pedestrians as they transition in different directions of movement affects the speed. Furthermore, We investigate the correlation between the rate of speed change as pedestrians transition in different directions and through different spaces. We regard our results can benefit the realistic pedestrian walking simulations, behaviour prediction, and speed based knowledge.