A research team from the Indian Institute of Technology Bombay (IIT-B), in collaboration with Zonal Railways and the Centre for Railway Information Systems (CRIS), has developed an innovative scheduling method to enhance the efficiency of Indian Railways. Their approach, termed ‘dailyzing,’ focuses on optimising train schedules without requiring extensive infrastructure upgrades.
The study, led by Prof Madhu Belur from the Department of Electrical Engineering and Prof Narayan Rangaraj from the Department of Industrial Engineering and Operations Research at IIT-B, addresses inefficiencies in non-daily train scheduling. While daily trains operate consistently, non-daily trains—those running only on specific days—are often scattered across the timetable, creating scheduling conflicts and underutilised railway sections.
The researchers proposed a clustering approach that groups similar non-daily trains running on different days into a single, predictable pattern. By treating these trains as a single unit within a structured 24-hour schedule, planners can minimise bottlenecks, reduce delays, and improve track utilisation.
To implement this, the team employed Hierarchical Agglomerative Clustering (HAC), a data-driven method that analyses train schedules to create efficient clusters. The technique proved highly effective in organising non-daily train operations, significantly reducing scheduling conflicts within the Indian Railways’ extensive network. Testing the model on India's Golden Quadrilateral and Diagonals (GQD) network, which links major cities like Delhi, Mumbai, Chennai, and Kolkata, demonstrated that HAC-based clustering could generate conflict-free schedules in seconds.
Prof Belur noted that the Indian Railways has already begun incorporating aspects of this model into its timetabling process on the GQD network. The implementation of an automated tool based on clustering has enabled railway authorities to optimise train movements and potentially introduce additional services without straining existing infrastructure.
Despite its success, the researchers acknowledge that scaling the model presents challenges. Long-distance trains frequently traverse multiple congested sections, requiring adjustments to clustering strategies at different points in their journeys. Additionally, as railway zones currently manage their timetables independently, better coordination across zones will be essential for maximising the benefits of dailyzing.
The study highlights how advanced data-driven methods can revolutionise railway operations without costly physical upgrades, potentially serving as a model for railway networks worldwide.