Analysis and Prediction of Pothole Formation Rate Using Spatial Density Measurements and Pavement Condition Indicators

Potholes represent one of the dangerous distress types on roads. They form as a result of various factors, including water ingress, freeze-thaw, and pavement condition deterioration. In the UK, 1.7 million potholes were repaired in 2021; this critical number causes significant economic, social, and environmental impacts; there is no tool able to predict the number of potholes that might appear in a road network. This study aims to analyze pothole formation and its relationship with other distress types and their severity, and to develop a simple tool able to predict the number of potholes that might appear in a road network based on the network condition. Significant pothole data from the road network of Greater London between 2017 and 2020, in addition to surface distress data, were used in this study. 'Spatial density' and 'join' tools embedded in ArcGIS were used to correlate pothole spatial density (PSD) with the road condition surrounding the potholes. This analysis was then used to calculate PSD as a function of different condition indicators, such as road condition index and crack intensity, allowing prediction of the number of potholes based on the length and condition of the sections being analyzed. The results demonstrate that potholes are significantly concentrated in sections with deteriorated conditions. They also show that it is possible to predict the number of potholes using PSD with reasonable accuracy. Lastly, it was found that sections with low crossfall are more susceptible to pothole formation, presumably because of water ponding and consequent damage.

Automated summary

This study analyzes the formation of potholes and their relationship with other distress types and severity, and develops a simple tool to predict the number of potholes that might appear in a road network based on the network condition. The results demonstrate that potholes are significantly concentrated in sections with deteriorated conditions and it is possible to predict the number of potholes using spatial density as a function of different condition indicators.