The Impact of Weather Phenomena on Passenger Volumes for Commuter Trains

Weather impacts several aspects of our daily life, of which the way we travel is one. The relationship between weather phenomena and ridership of trains has not received much attention previously. The study we present here aims to understand the impact of temperature, wind, and precipitation on passenger volumes for commuter trains. To do so we make use of automatic passenger count and weather data from over a million unique station stops, spanning two years, in the Southern region of Scania in Sweden. Our findings show that changes in the level of precipitation do not affect the volume of boarding passengers. Statistically significant effects are found for changes in temperature and wind speeds. These effects are most prominent for departures outside of peak hours. The results are useful for planning more accurate dwell times and rolling stock circulations and can serve as inputs during real-time rescheduling problems and demand modelling