The Story of a Chart: Passenger Flows on the Orange Line

During the June 6 State of the Orange Line presentation (PDF) at the MBTA Fiscal and Management Control Board meeting, the MBTA presented this chart which shows the usual load at each station on the Orange Line during peak service.

Chart from the FMCB Presentation depicting average passenger flows by stop on the Orange Line

This chart shows the average number of passengers “flowing through” each station during 30 minutes of the morning peak (8:00-8:30 AM) on the Orange Line. This includes all the passengers who enter the station during that time period as well as the passengers who are already on trains when they arrive at that station. These data are estimated from a combination of station entries as recorded by the AFC system, and exits and transfers as estimated from the ODX model.

The calculation is relatively simple:
(existing load) + (passenger entries at each station) - (exits) = (passenger load)

The above table shows a portion of the raw data to illustrate this. Heading northbound, an average of 1,366 (rounded) people get on trains at Forest Hills during this time period. That makes the flow at this point (obviously) 1,366. Then, at Green Street, an additional 402 people board trains, and 14 get off. That makes the flow from Green Street to Stony Brook:
1,366 + 402 — 14 = 1,754

Repeating this for each stop along the route gives us the “flow” of passengers.

It may be easier to visualize this for just one train — a load per train — but we believe it useful to show the total flow of people through the corridor, rather than just one train. To estimate the load for each train, you could divide these numbers by 5.

The dotted line on the chart illustrates the theoretical capacity of the system during this period. This is calculated by the following formula:
(policy capacity of each car) × (cars per train) × (number of trains per time period)

In this case, the policy capacity of an Orange Line car is 131 people, there are 6 cars per train and 5 trains arrive each half hour, making the capacity of the Orange Line during this time period:
131 × 6 × 5 = 3,930

Some assumptions made here are important. They affect both the flows and the capacity as shown in the chart:

  • The amount of passengers at each station used in this chart is an average and this can and does vary. While we have already filtered out weekends and holidays, Fridays are usually less busy, and sometimes there is higher or lower demand for various reasons. There is also variability in demand within the 30-minute period illustrated.
  • The capacity of each train car used here is what is known as a “policy capacity,” which is simply a calculation based on the amount of floor space and number of seats in the car. This is meant to be a capacity in which all passengers feel comfortable, as we know more people can actually fit on each car. This number assumes both equal distribution of passengers throughout the car and that each passenger takes up the same amount of space. For more on crowding and passengers’ tolerance of it, see At What Level Does Bus Crowding Become Unacceptable?
  • We also assume passengers distribute themselves evenly on each platform and in each car.
  • Headways between trains are assumed to be consistent.

As you can guess, these assumptions are rarely all true at once, making the capacity line mostly theoretical. But it does accurately convey the point that the Orange line is crowded during peak times, and illustrates the places along the line where it is most crowded. Additionally, using the entry and exit data, we can accurately identify locations with the most passengers both boarding or alighting trains and flowing through stations and try to plan better for these areas of crowding.