MUMBAI, India, June 22 -- Intellectual Property India has published a patent application (202621048354 A) filed by Moresh Madhukar Mukhedkar; Sanskar Jagpal Tolumbia; Devang Rushikumar Arekar; Priyanshu Tejas Tamboli; Ayush Deepak Choudhary; Vishal Patil; and Shweta Koparde on April 16, 2026, for Ai-Based Intelligent Train Traffic Control System.
Inventors include Moresh Madhukar Mukhedkar; Sanskar Jagpal Tolumbia; Devang Rushikumar Arekar; Priyanshu Tejas Tamboli; Ayush Deepak Choudhary; Vishal Patil; and Shweta Koparde.
The application for the patent was published on June 12, 2026, under issue no. 24/2026.
Abstract: The present invention relates to an intelligent system called RailOptima that enables railway junction controllers to manage platform assignment, train priority queuing, and conflict detection through an AI- driven simulation and decision-support dashboard. The system integrates constraint-based optimization using Google OR-Tools CP-SAT, a multi-factor priority scoring engine, and a Random Forest Regressor delay prediction model to automate and assist operational decisions at high- traffic rail junctions. It consists of a Python FastAPI backend with a simulation loop powered by APScheduler, a SQLite persistence layer managed through SQLAlchemy ORM, and a React and Vite frontend connected via WebSocket for real-time data delivery. The system assigns platforms to waiting trains subject to availability and compatibility constraints, scores trains by priority, detects platform conflicts, and generates human-readable recommendations for controllers. Testing via simulation runs using a seeded Mumbai CST to Pune Junction timetable confirmed the correctness of the priority engine, optimization cycles, and conflict detection logic. By replacing manual decision-making with quantitative, data-driven automation, the invention improves consistency, reduces delay propagation, and enhances the operational capability of railway junction management.
Disclaimer: Curated by HT Syndication.
