Interrogating the Spatio-Temporal Dynamics of Urban Footfall using Principal Component Analysis

As more and more of the world’s population move to urban areas, understanding pedestrian dynamics becomes increasingly important. Data sources that measure pedestrian counts offer great potential as a means of interrogating urban flows without needing to track individuals. This study attempts to better understand the fundamental spatio-temporal structures that characterise aggregate pedestrian behaviour. It employs Principal Component Analysis to distill the key temporal elements that comprise aggregate daily pedestrian activity across the city. We identify clear, distinct components that represent different types of activities and can be used to characterise different types of neighbourhood.