A Global Modeling Pruning Ensemble Stacking with Deep Learning and Neural Network Meta-Learner for Passenger Train Delay Prediction

Train Operators can improve railway passengers’ service quality and traffic management by accurately predicting travel arrangements delays. Precise prediction of train delays is vital for creating feasible scheduled timetables. The import pruning stacked ensemble deep neural networks into delay helps model accuracy computational time. In this study, we propose a novel learning that uses pruned multilayer perceptron (MLP) as meta-learner heterogeneous sub-models to the passenger arrival evaluate performance using Amtrak, most extensive US railroad data, determine enhanced accuracy. We used generalization regression artificial create models. optimized models OPTUNA produced best network an MLP meta-learner. Our methodology comprises data preprocessing, feature engineering, modeling, case studies, evaluation phases. experiments demonstrate our proposed Neural Network (PST-NN) has various degrees improvement in indicators compared existing 35.20%, 53.40% terms error outperforms benchmark 85.22% Passenger respectively, which provides new approach building efficiently prediction.