Relationship between Pavement Roughness and Distress Parameters for Indian Highways: Fid-Bau Portal

Relationship between Pavement Roughness and Distress Parameters for Indian Highways

Chandra, Satish / Sekhar, Chalumuri Ravi / Bharti, Anish Kumar / Kangadurai, B.

The present study demonstrates the relationships between pavement roughness and distress parameters like potholes, raveling, rut depth, cracked areas, and patch work. The pavement distress data collected on four national highways in India using a network survey vehicle (NSV) are used to develop linear and nonlinear regression models between roughness and distress parameters. Analysis of variance of these models indicated that nonlinear relation is better than a linear model. $R^{2}$ value, root mean square error (RMSE), and mean absolute relative error (MARE) also supported nonlinear models. An artificial neural network (ANN), which is an advanced technique of modeling, is also used in the present study to model pavement roughness with distress parameters. A network with five input nodes, 15 hidden nodes, and one output node is considered. The network was trained with 90% of the data and tested with remaining 10% data. Results of $R^{2}$ and MSE showed that the neural network performed highly significantly in both training and testing phases. Finally, the performance of the ANN model is compared with that of linear and nonlinear regression models. The mean absolute error (MAE) for the ANN model is around 18% less than that for the linear model and 11% less than that for the nonlinear model. MARE values are also 12.5% lower in the case of ANN modeling, indicating that the ANN model yields a better forecast of road roughness for a given set of distress parameters.