This research focuses on enhancing Sri Lanka’s railway system through a Real-Time Train Tracking and Time Prediction System, tailored to the country’s unique railway network. The system integrates advanced technologies like machine learning and GPS for precise time predictions, which are communicated via a website. The study begins with an in-depth analysis of Sri Lanka’s existing railway infrastructure and operational complexities. Train NO. 1015, which travels daily from Colombo to Badulla, was selected for this research. Factors such as weather, passenger delays, operational requirements, signaling errors, and staff delays were identified as challenges to accurate time predictions through an extensive data analysis of past train delay data. The system design includes a machine learning algorithm using LSTM (Long Short-Term Memory) neural networks for predicting train arrival times, and a user-friendly interface for displaying these predictions. The goal is to provide real-time insights for railway authorities and reliable travel information for passengers through a user-friendly web application.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Real-Time Train Tracking and Time Prediction System For Railway System in Sri Lanka




    Publication date :

    2024-08-23


    Size :

    939866 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English




    TRAIN AND RAILWAY SYSTEM

    NOGI MASAYUKI / SATAKE NOBUHIKO | European Patent Office | 2015

    Free access

    Railway train overhaul system

    WEI LAI | European Patent Office | 2015

    Free access

    Train operation parameter real-time monitoring system

    ZHANG RUIXUE / WANG WEI / SONG LILI | European Patent Office | 2015

    Free access

    Modeling and Solving Real-Time Train Rescheduling Problems in Railway Bottleneck Sections

    Chen, Lei / Roberts, Clive / Schmid, Felix et al. | IEEE | 2015