Train Delay Prediction using Machine Learning


Delay prediction is a process of estimating delay probability based on known data at a given checkpoint and is typically measured via arrival (departure) delay. The key to making delay predictions based on actual operational data involves establishing the relationship between train delays and various characteristics of a railway system.This provides a basis for the operator’s scheduling decision Train delay is a significant problem that negatively impacts the railway industry and costs billions of dollars each year. In this project we have used Train delay dataset from IRTC to predict Train delays.  We have used Faster RCNN algorithm to predict flight departure delay and our model can identify which features were more important when predicting Train delays.


Traditional statistical machine learning methods consider train operation time features as model data to update algorithm structure and parameters in time such as delay prediction using Decision Tree Algorithm, SVM, LSTM Algorithms. The actual delay data of the train should be updated in real time, But its Failed to Update on Real Time.


Bi-LSTM, SVM Algorithm cannot obtain precise results because of less features Considered.

SVM are prone to errors.




Proposed Delay prediction is a process of estimating the probability of train delays at subsequent recording points based on train operation history data, and this is typically determined by arrival delays.

  • The train can be delayed due to various disturbances in the operation process.
  • Six parameters are selected after the analysis of the train arrival delays at the station to constitute the feature space (F).
  • Train Characteristic(X1)
  • which day of the week it was (Monday, Tuesday etc., one attribute for each)
  • Whether is was weekend (Saturday or Sunday),
  • Time of the day (one for morning/evening and one for day/night)
  •  Weather attributes (temperature, snow depth, wind speed, visibility),
  • Train line the departure happened on (X0, X1, X3).

Software Requirements:

  • Front End – Anaconda IDE
  • Backend – SQL
  • Language – Python 3.8

Hardware Requirements:

  • Hard Disk: Greater than 500 GB
  • RAM: Greater than 4 GB
  • Processor: I3 and Above


    train delay prediction using machine learning, forecasting and analysis of train delays and impact of weather data using machine learning, train delay prediction neural network, predictions of train delays using machine learning, railway tracking and arrival time prediction system project, railway enquiry arrival time, train delay prediction system a big data analytics perspective, train delay prediction using machine learning project code github, Ieee machine learning projects 2023, ieee python projects 2023