Train Delay Prediction Using Support Vector Machine

Abstract

Passenger train delay is one of the essential problems in railway systems across the world. Train delay is a major problem in the transportation sector. Poor on time performance can impact passengers' trust and it may result in a shift to other modes transportation. Due to poor signaling and less number of railway tracks, Trains are unable to run on their scheduled time. Support vector machines (SVM) algorithm is used for analyzing or forecasting the train delay. To extract features, principal component analysis is employed. A statistical method for decreasing a dataset's dimensionality is PCA. The travelers would benefit more if their journey could be planned around their jobs. By predicting the train delay (on-time or before or late) performance in terms of time and train, the travelling plan will be rescheduled.