Passenger arrival rates at public transport stations

Abstract

The amount of time spent waiting at a public transport station is a key element in a passenger’s assessment of service quality and in mode choice decisions. Many transport models estimate the average wait time is half the headway for small headways and use a maximum waiting time for headways over a given value. The assumption is that at small headways passengers do not bother to consult schedules since vehicles arrive frequently; therefore these passengers arrive regularly at the station. In contrast, at longer headways passengers do consult schedules to reduce their waiting time; these passengers arrive clustered around the departure time. This research evaluated the influence of headway and other factors on passenger arrival rates at public transport stations based on data collected at 28 stations in Zurich’s public transport network. It found that even at 5-minute headways, some passengers consulted schedules and did not arrive randomly at the station. This finding is interesting since 5-minutes is much lower than many models assume, therefore these models may be overstating passenger wait time. The research also found time-of-day and reliability had an important influence on passenger arrival rates. The research proposes a model for passenger arrival rates at stations that combines a uniform distribution with a shifted Johnson SB distribution.