IIT Bombay and CRIS develop innovative system to improve train scheduling

To address these issues, the team developed the concept of ‘dailyzing,’ which clusters non-daily trains based on similarities in their schedules. Prof Madhu Belur, from the Department of Electrical Engineering at IIT Bombay, explained that dailyzing involves grouping non-daily trains into clusters, making them operate as if they were daily services. “This process groups trains that share similar routes and times—within a 15-minute window—into predictable patterns. According to researchers, by treating these non-daily trains as part of a structured timetable, rather than as independent entities, railway planners can reduce inefficiencies and optimize track usage. This results in a more efficient timetable, filling scheduling gaps and allowing better resource utilization,” he said.

Prof Belur was joined by Prof Narayan Rangaraj from the Department of Industrial Engineering and Operations Research at IIT Bombay, and experts from Zonal Railways and CRIS. The team tested their model on India’s Golden Quadrilateral and Diagonals (GQD) network, a major rail system connecting Delhi, Mumbai, Chennai, and Kolkata. They used clustering techniques like Hierarchical Agglomerative Clustering (HAC), Density-Based Spatial Clustering of Applications with Noise (DBSCAN), and K-means to analyze real-world train data.

The researchers found out that the clustering approach has led to more compact and efficient timetables. It could also free up space for additional services. “For instance, if a cluster has fewer than seven trains (one for each weekday), extra trains can be added on free days, improving scheduling flexibility. The researchers believe this could help manage bottleneck sections more effectively, reduce delays, and optimize train flow,” said Prof Belur.

He said that Indian Railways, with support from the research team, has already begun implementing a modified version of the dailyzing model to improve scheduling on the GQD network and further refinements, such as integrating real-time adjustments, which could make the system even more adaptive and seamless, enhancing the overall efficiency of India’s vast railway network.