Citation

Evans J, Waterson B, Hamilton A. J. Transp. Eng. A: Systems 2020; 146(6): e03120001.

Copyright

(Copyright © 2020, American Society of Civil Engineers)

DOI

10.1061/JTEPBS.0000362

PMID

unavailable

Abstract

Only a small minority of road incident detection algorithms (IDAs) have been designed for use in urban road networks. A review of the literature conducted by the authors revealed that approximately 10% of published papers on novel IDAs are designed specifically for urban networks. Urban networks present many challenges that are not faced on highways, such as signalized junctions causing queuing similar to that of an incident. This paper reviews the progress made in urban incident detection research and highlights where improvements are needed. It is found that few algorithms have been implemented or tested on real-world data, and so few have been sufficiently evaluated or compared. Those that have been implemented often find difficulty in differentiating disruption from incidents and contextual factors such as sporting events or public holidays. This has caused unnecessary false alerts to be triggered, leading to dissatisfied operators. Progress could be made if more research data and IDAs were published, allowing for thorough evaluations and comparisons of algorithm performance on real-world data. Further research is required to improve IDAs' ability to differentiate between incidents and contexts, particularly in complex urban networks.

Language: en