In this paper, we explore a novel approach for identifying critical congestion hotspots based on network analysis using the Automatic Number Plate Recognition (ANPR) data in Cambridge, UK. We transform the camera surveillance data into undirected and weighted travel networks and discuss the smart placement of camera sensors based on this case study. Specifically, the paper addresses two questions: (a) How can we use conceptualized travel networks to reveal congestion patterns? (b) Is the placement of the camera sensors efficient and optimal in the case study area? The results show that: (1) The travel network model can effectively identify critical congestion points. Using only 44% of the total cameras, we accurately capture the major congestion hotspots, including the junctions around the urban fringe. (2) The current layout of the camera sensors is redundant in several local areas. We therefore suggest that spatial optimization should be considered for sensor placement in the future. This study sheds new light on revealing urban mobility patterns in relation to travel networks and provide a heuristic tool for decision-makers for advancing smart digitalization in their cities.

CICTP