An Empirical Study of Timetable Strategies and Their Effects on Punctuality

Punctuality in Sweden has been too low for several years, and there is now a goal in the industry that it must improve by more than five percentage points by 2020. Research has shown that timetable properties affect punctuality, and timetables can be changed relatively quickly and easily, compared to infrastructure and rolling stock. In this paper we study some strategies for assigning margins in timetables, and how they affect punctuality. The aim is to identify changes that can improve the punctuality of passenger trains. The study is based on a dataset containing detailed timetables and records of train movements for all trains in Sweden during the year of 2015, over 1.1 million departures. We find that every additional percentage point of margins improves punctuality by about 0.1 percentage points, and about the same for every percentage point increase in the weighted average distance of margins. It is common for margins on some line sections to be negative, which lowers punctuality by on average 4.1 percentage points. Small time supplements directly following station stops are associated with a punctuality drop of 2.5 percentage points. We also find that the number of interactions between trains at stations, and on line sections, lowers punctuality by 1.2 and 3.9 percentage points per instance, respectively. From these results we propose a package of changes to timetable planning which we estimate would improve overall punctuality by up to five percentage points