Joe Matthews
Early detection of collision hotspots.
Email: [email protected]
Supervisors
- School of Engineering
- School of Mathematics, Statistics and Physics
Project description
Improving road safety is a global challenge. Road safety practitioners are becoming increasingly reliant on data to find solutions. Problems can arise in cases where data are limited. In such cases, we must account for confounding factors such as Regression To the Mean (RTM) to avoid misleading results.
We are investigating methods for road safety hotspot predictions. We are developing a model which proactively identifies future hotspots, enabling early treatment.
Our model combines network-wide covariate data and allows for local site effects and trends. We extend this to include seasonal data. We then fit structural models to the seasonal and spatial effects to reduce uncertainty in estimates. This also allows for interpolation between sites.
We add components to estimate collision severity and/or collision type factors. This again models network-wide and local effects. Thus, it allows for proactive detection of collision hotspots.
Publications
- Matthews JT, Newman K, Green AC, Fawcett L, Thorpe N, Kremer K. A decision support toolkit to inform road safety investment decisions. Municipal Engineer 2019, 172(1), 53-67.
- Fawcett L, Thorpe N, Matthews J, Kremer K. A novel Bayesian hierarchical model for road safety hotspot prediction. Accident Analysis & Prevention 2017, 99(Part A), 262-271.
- Fawcett L, Thorpe N, Galatioto F, Slater P, Hoffmann T, Kremer K, Muench A. Identifying Collision Hotspots using Time Series Analysis and Accounting for Regression to Mean. Paper presented at the 13th Annual Transport Practitioners’ Meeting. Session: Road Safety: Design Applications. 2015. UK: London.
Qualifications
- MMathStat, Newcastle University