Intelligent Assessment of Pavement Condition Indices Using Artificial Neural Networks

The traditional manual approach of pavement condition evaluation is being replaced by more sophisticated automated vehicle systems. Although these systems have eased and hastened management processes, research ongoing to further improve their performances. An average state road agency handles thousands kilometers the network, most which multiple lanes. Yet, for practical reasons, are designed evaluate networks one lane at a time. This requires time, energy, possibly equipment manpower. Multiple Linear Regression (MLR) analysis Artificial Neural Network (ANN) were employed examine feasibility modeling predicting distresses lanes as functions single adjacent lane. successful implementation this technique has potential cut energy time requirement stage least half, uniform multi-lane highway. Results showed promising model performances that indicate possibility evaluating highway (PC) inspection. Traffic direction parameters, location, matching parameters contributed significantly performance ANN PC prediction models.