Vehicle Classification to See the Effect of Damage on Flexible Pavement

According to the National Institute of Statistics and Informatics (INEI) in the technical report N°5 - May 2023, the number of light and heavy vehicles through toll booths grew by 2.4%, resulting in traffic congestion and pavement damage due to the lack of adequate vehicle counting control. This work developed a vehicle classification system to assess damage to the flexible pavement of the Universidad del Centro del Perú (UNCP) roads. For the vehicle classification system programming, the registration area was identified using Google Earth Pro. Subsequently, the vehicle categories to be classified were established: car, pickup truck, bus, and trailer. These vehicles were trained using a neural network over 50 iterations, with Python utilized to develop the vehicle counting and classification algorithm. While to assess the pavement damage, the data obtained from the field of manual counting and simulation was required to obtain the equivalent factor that is a function of the type of axle and weight of each vehicle. Finally, there was a 19% deficiency in the vehicle classification of cars and the damage to the flexible pavement was determined by considering the manual count and the programmed system count, for which in both cases it was found that the vehicle that damages the flexible pavement is the cars, with 23.22% of damage in the one-way lane in the manual count and 49.49% in the system count.