Application of Machine Learning Algorithms for Railway Queue Analysis in Tamil Nadu

Abstract: This study investigates the operational efficiency and queuing behavior of Southern Railwaystations in Tamil Nadu using machine learning techniques. The analysis is based on a secondary datasetcapturing key operational parameters, including Arrival Rate, Service Rate, Queue Length, WaitingTime, and Throughput Index, collected from multiple stations. The study employs three machinelearning models are Logistic Regression, Support Vector Machine (SVM) with RBF kernel, and NeuralNetwork (MLP)to predict the operational Level of each station, categorized as Normal, Warning, or Critical. Data preprocessing, feature engineering, and normalization were performed to ensure high-quality inputs for model training. Model performance was evaluated using metrics such as accuracy, precision, recall, F1-score, and ROC-AUC, while visualization tools, including Confusion Matrices,ROC curves, and Clustered Bar Charts, were used to interpret and compare the predictive results.Among the models, Logistic Regression achieved the highest accuracy of 97%, followed closely bySVM at 96.5%, while Neural Network reached 89% accuracy. The results demonstrate that machinelearning algorithms can effectively predict congestion levels and support informed decision-making forrailway operations. The study highlights the potential of integrating predictive analytics with railwaymanagement to optimize queue handling and improve passenger satisfaction.