The escalator strikes back

Publish date:

Exactly a year ago, Lukas Dobrovsky and I wrote a blog post in response to an experiment conducted by TfL, in which they investigated the effects of standing on both sides of escalators on commuters travel time at Holborn Station.


Figure 1: How do an escalator right (and left!)

The blog was well received, however, things have escalated recently as the subject was brought back into the spotlight after the New York Times wrote an article in which Paul Wiedefeld, the general manager of the Metro in Washington D.C., was quoted as saying that “the practice of walking on the left and standing to the right — as outlined in the Metro’s rules and manners could damage the escalator.” (New York Times)

This has led to an incredible amount of interest in the original piece and has resulted in it being featured in multiple forms of press and media, including interviews and articles in; NBC Washington, The New York Times, The Independent, ABC in Australia and FOX 5 on US TV.

Therefore, as with all great movie franchises, we thought the original research needed a sequel!

So, just to recap, on an escalator, commuters either stand on the right or walk up the left hand side. Those who stand on the right-hand side tend to leave every second step empty while those who walk leave up to three steps between each person, which wastes even more space. Moreover, the majority of travellers prefer to stand, particularly in stations with long escalators, which further decreases the utilisation of the left-hand side.

The blogging team built a simulation model in 2016 to mimic the real life behaviour of London Underground commuters using an escalator during the rush hour period on a Monday morning.

The original study, which assumed all commuters were standing, produced some interesting findings (see the original blog for more details). There was a dramatic decrease in the total time taken, from queuing to get on the escalator at the bottom to stepping off at the top, of 79 seconds for those who originally stood on the escalator anyway (Figure 2).

Figure 2: Average time to the top of the escalator

A further effect observed while using the simulated model was that the queue to get on the escalator fell from an average of 73 to 24 people, showing additional benefits for those who like to stand. There were also similarly positive results in the TfL experiment: “The standing-only escalators were able to carry an average of 151 passengers per minute, compared to 115 on the ‘normal’ escalator, where people were both walking and standing. So, the theory was correct — having people stand on both sides of the escalator does increase capacity” (Londonist). Despite this, and the success of the pilot, the scheme has not been rolled out across London.

Due to the recent publicity, the blogging team have had a lot of feedback on the original piece, especially from people who would normally choose to walk up the left-hand side of escalators. The general view from this group of people was negative which suggests they would be strongly against a policy such as this being enforced, as they largely disagree with the implied results. As a result, the team have run two new simulations, using the original Simul8 simulation model from 2016, to test some different scenarios.

Figure 3: Results of Scenario 1 and 2

Scenario 1 was suggested by some of our readers. It should be noted that it would be impossible to enforce this. What is interesting is that there would still be a queue to get on the escalator as a result of wasted capacity (not all steps are used). On average, the queue length would be 26 people over the course of the 100-day trial during rush hour. However, the time in the system would be just 46 seconds.

So, what would happen in Scenario 2? This would not be a viable option for a lot of underground stations due to the number of escalators required. For the purposes of the experiment we doubled the size of the population, given that only the busier, large stations are likely to have more than one escalator to make this feasible. Variance was still included in the model to ensure factors such as people’s different walking speeds were accounted for.

Drum roll, please! The results were as follows:

Figure 4: Results of Scenario 2

Even though the number of commuters has doubled, there are clear benefits of separating walking and standing commuters. Compared to the results of the original blog (bearing in mind that the population sizes are different), walkers would gain an extra six seconds and have shorter queues to get onto the escalator and they can still stack up their steps on their Fitbit. Those who stand would move through the system quicker by 71 seconds and average queue lengths would fall by 35 people to 38.

The potential for using simulation models such as this to test further theories is huge and, compared to the 3-month experiment conducted in real life, could save both time and money. There are many more scenarios that could have been run within a matter of minutes to replicate similar experiments, whether they are related to the original topic or have an entirely new subject matter.


Check out previous blogs that Shivam has previously contributed to:

  • The NHS missed their waiting time targets – an accident or an emergency? (Link)
  • It could be you, but it probably won’t be. (Link)
  • The price of misery (Link)

By continuing to navigate on this website, you accept the use of cookies.

For more information and to change the setting of cookies on your computer, please read our Privacy Policy.


Close cookie information