Computer-aided prediction of Pavement Condition Index (PCI) using ANN
Conference Paper
Overview
Identity
View All
Overview
abstract
The accurate prediction of pavement network status and performance is essential for development of pavement Maintenance and Rehabilitation (M&R) Plans and competent management of the transportation infrastructure system. Hence, pavement conditions need to be monitored and evaluated properly, so that adequate conditions, and, consequently, safety and comfort can be ensured during the entire road service life. With this respect, Pavement Condition Index (PCI), an indicator of great value in pavement engineering, which rates the surface condition and the structural integrity, is used in this research. In this paper, an optimized artificial neural network (ANN) model has been used to predict the PCI based on the experimental study carried out on Texas A&M University campus. A multi-step approach has been carried out to find the optimized ANN model to predict PCI founded on the experimental data obtained. The results show that for PCI prediction, which can be considered as a complex civil engineering and management problem, the accuracy of optimal ANN based on this approach is well above the normal ANN models.
