Real-Time Passenger Train Delay Prediction Using Machine Learning: A Case Study with Amtrak Passenger Train Routes

Abstract

The traveller train delays have a substantial impact on users’ decision to use rail transit. Using methods of machine learning, this paper provides real-time passenger train delay prediction (PTDP) models. The influence on PTPD models employing Real-time based Data-frame Structure (RT-DFS) and Real-time with Historical based Data-frame Structure (RWH-DFS) is looked at in this essay. The outcomes prove that PTDP models that combine MLP and RWH-DFS outperform all other models. The result of external variables such as historical delay profiles at the destination (HDPD), ridership and population, day of the week, geography, and weather data on real-time PTPD models is also examined and explored. This system’s ability to improve the precision of anticipating train arrival delay time is critical for airport improvement.
Transportation effectiveness.In our procedure, we must use to increase the accuracy of dataset as input. Following that, we must incorporate using machine learning like logistic regression and random forest. The experimental findings reveal that each algorithm’s accuracy and error values are different. The model has a high forecast accuracy and can accurately follow the trends of several delay indicators.