Analysis of Flexible Pavement Serviceability Using ANN for Urban Roads

Serviceability is an indicator that represents the level of service a pavement provides to the users. This subjective opinion is closely related to objective aspects, which can be measured on the pavement's surface. Modelling the present serviceability index is very important in pavement management systems. In the present study, the present serviceability index (PSI) has been analyzed using artificial neural networks (ANN) for the flexible pavements of urban roads. The study area considered constitutes 21 urban road sections for Noida city in the NCR of New Delhi, capital of India. Field data collected include slope variance, rut depth, patches, cracking, and longitudinal cracking for two consecutive years on the selected road network. The developed ANN model was also compared with a model developed using multilinear regression analysis.