Data-driven public transport schedule refinement: a case study of the tram network in Budapest

In many cities, urban rail networks form the backbone of public transport. Since this infrastructure is already in place, there is limited room for major structural changes. Timetable optimization offers a cost-effective way to improve service quality. This paper presents a data-driven method for refining Budapest’s tram schedules using trip-planning queries from the local travel application as a proxy for intended demand. The methodology iteratively replans users’ intended trips with OpenTripPlanner and adjusts tram departure times using a genetic algorithm to improve the schedule. Results show that small timetable offsets can reduce travel time and measurably influence passenger routing. The findings underline the necessity of analyzing schedules in a network-wide context rather than optimizing isolated lines. The methodology is generalizable to other urban contexts and directly actionable without significant infrastructure investment.