Predictive System to Avoid Passenger Saturation on the Lima BRT using Neural Networks

Mellarez A.; Teodoro G.; Antezana R.; Cynthya A.; Flores M.; Elias R. · Proceedings - 2025 3rd International Conference on Intelligent Control and Computing, IC and C 2025 (2025) · DOI: 10.1109/ic-c68228.2025.00026

Resumen

The main cities in the world have a transport service that allows citizens to move around them. Lima has a public service called Bus Rapid Transit (Lima's BRT), which allows the transportation of citizens. This service presents situations of passenger accumulation in all its stations caused by the poor determination of supply and demand of buses; this results in delays in departures and journeys, overloading of buses and the annoyance of people, so the quality of the service is diminished without fulfilling the objective of rapid transport. In this research, a predictive model has been developed using neural networks that allows estimating the flow of passengers in the stations, providing the competent authority with an important resource for determining the demand and supply of buses with the objective of speeding up the flow of passengers. The model allows a precision of 92% compared to other models and it has been proven that the estimation of the use of buses allows speeding up the flow of passengers, satisfactorily speeding up the service.

Instituciones: Universidad Peruana de Ciencias Aplicadas