Where a train stops, it helps to know

Does it really matter at what point along a platform a passenger train stops, or “spots,” to board and alight riders? According to a Dutch researcher who compared rider behavior when trains were stopped at different points on the platform at Amsterdam’s Schiphol Airport station (one of the Netherlands’ busiest), it matters a great deal.

Most of the Netherlands’ long-distance and regional trains run through Schiphol, with longer-distance trains running once or twice hourly and regionals running six to ten times hourly. The 40-year-old station is enclosed under the airport, and is always crowded, with a great percentage of passengers carrying luggage. It has six tracks with three island platforms. There is no room to add tracks or lengthen the platforms, though NS, the Dutch national railway (not to be confused with Norfolk Southern!), anticipates a doubling of station patronage (currently 85,000 daily passengers) in the next 15 years.

Reliable schedules and dwell times are crucial to keeping the flow of passengers in and out of the station fluid. Regionals, consisting of four single-level low-floor coaches, dwell for one minute, while intercity trains, made up of twelve bilevel coaches, dwell for two. Platforms are 90 centimeters (just over 35 inches) high. Researcher Jeroen Van den Heuvel, a Ph.D. student in Station Engineering at Delft University working under contract to NS Stations, measured the amount of time a train occupies a given track section (in the station or for trains waiting just outside of it for a train in front to clear the way) for a sample of 300 train departures as a proxy for how efficiently passengers were loading and unloading. He controlled for “biased cases,” instances where a train was delayed or there was a track change or other disruption, ending up with 200 “unbiased” departures.

Among Van den Heuvel’s findings were that the platform position of the train (based on the placement of a sign above the overhead wire telling the engineer where to stop) had no significant effect on dwell time when ridership was light. But on trains and at times with heavy ridership, dwell time averaged about 20 seconds higher overall than at lighter times. A better positioning of the train on the platform with respect to the locations of escalators and elevators, and accounting for typical traveler behavior, cut the dwell time by about 30 seconds at the busiest times — which is significant for a tightly-scheduled operation. Multiply that time savings by the number of trains every hour, and the proper platform positioning could allow the railroad to shave minutes off its schedules. In the case of Schiphol station, he found it best to stop the train closer to the escalators at one end of the platform, even though the platform is narrower there.

Van den Heuvel also found qualitative differences between level boarding (where the train’s floor is roughly the same height as the platform) and passengers having to step up to board a train. He also suggested further study is needed to compare situations where passengers have a lot of luggage to those where they tend to travel lighter. He also noted that NS’s wheelchair-bound passengers must call the railroad one hour in advance of departure so that staff can be made available to put movable wheelchair plates in place. (Such a burden on passengers would violates the purpose of the Americans with Disabilities Act in the US, which is to allow passengers with disabilities to board freely at any time independently, without crew assistance.)

Although Van den Heuvel did not study any other stations in-depth, he observed that for stations that are outdoors or not completely under a cover, it is important to either have a canopy cover the entire platform, or for the train to stop so that as many boarding doors as possible are at parts of the platform covered by the canopy. He also said station designers should consider the strategic placement of benches or leaning posts along a platform, and that even spacing of access points (doors, walkways, elevators, escalators or stairways) along the platform is preferable to having them all in one place, either in the middle or at the ends of the platform.

Of course, performing this kind of experiment and taking advantage of synergies between station layout and train operations was made much easier by the fact that NS has almost complete control over the station and all trains serving it. In the US, train operators (Amtrak in particular) often do not own or control the stations they serve; stations are served by multiple operators (intercity and commuter, for example), and station service procedures require negotiation between station management, the operator and the host railroad. There is also the matter of platforms that only serve one track on a double or triple-tracked line. But if all these stakeholders could successfully cooperate on even seemingly minor matters like this, significant benefits are possible to the fluidity of rail traffic and to customer satisfaction.