A novel approach for predicting urban pavement damage based on facility information: A case study of Beijing, China

This study aims to establish a novel approach for predicting urban pavement damage without using any road inspection data. For this purpose, three indicators are constructed to describe the road-use intensity based on the concept of centrality in transportation planning and economic geography. These indicators are defined to represent two variables: importance to the city and centrality for surrounding users. In addition, these indicators are adopted in a model for pavement damage prediction that follows a classic logit form, the application of which is illustrated for a selected sample area in Beijing. The significance and validity of each indicator are verified. The empirical results demonstrate that the proposed model is acceptable and achieves a prediction accuracy rate that is generally greater than 85%. Furthermore, the proposed approach is more cost-effective than traditional methods of pavement damage prediction.