Public transportation is a crucial component of urban mobility in cities. However, one of the main challenges that users face is the lack of accurate information about arrival times at each stop. This article presents a supervised learning approach using artificial neural networks for estimating travel time between any two bus stops. The dataset comprises information from 2 073 trips, incorporating various factors, e.g., starting and arrival times, speed, total travel time, the number of speed bumps and traffic lights, distance traveled, and other unexpected events. The case study focused on urban public transport in Cuernavaca, Mexico, covering the period from October 1, 2024 to February 20, 2025. Results showed that the proposed artificial neural network model achieved a high coefficient of determination, ranging from 0.9995 to 0.9998, compared to the baseline method for predicting arrival times. The proposed approach highlights the potential of using artificial neural networks as a decision-making tool to optimize transportation arrival times at each stop, thus facilitating better organization and planning of travel schedules.