In most road pavements design methods, a solution is required to transform the traffic spectrum to standard axle load with using equivalent axle load factor (EALF). The EALF depends on various parameters, but in existing design methods, only the axle type (single, tandem, and tridem) and pavement structure number were considered. Also, the EALF only determined for experimental axles and axle details (i.e., axle weight, length, pressure), wheel type (single or dual wheel) plus pavement properties were overlooked which may cause inaccuracy and unusable for the new axle. This paper presented a developed model based on Artificial Neural Network (ANN) for calculation of EALFs considering axle type, axle length, contact area, pavement structure number (SN), tire pressure, speed, and final serviceability. Backpropagation architecture was selected for the model for the EALF prediction based on fatigue criteria. Finally, among all reviewed ANN configuration, a network with 7-13-1 was selected for the optimum network.

Keywords