Using Pedestrian Modelling to forecast public transit capacity following COVID-19

As governments begin communicating plans to reopen our cities in the coming months, an important consideration will be how our transit infrastructure is able to accommodate the increasing passenger volume while prioritizing the health and well-being of commuters. Fear of contagion will likely continue for some time, and this will impact the behaviour of entire communities. Transit authorities who can accommodate people safely and comfortably, and who are able to articulate the steps they have taken to assess the capacity of their infrastructure during and following the coronavirus pandemic, will gain the public’s trust and play a crucial role in getting us all back to work.

Acknowledging that there is still much change ahead of us, we have been exploring how we can support our clients to make informed, agile decisions as they begin to contemplate life and work after COVID.

Here we have presented preliminary modelling and some thoughts on two hypothetical scenarios based on what we know today. In these scenarios, we have expanded the functionality of existing software using custom programming scripts to quantify performance metrics that have increased in relevance.

Is a socially distant commute possible?

We regularly examine how our clients might quantify the performance of their building and infrastructure as it relates to pedestrian movement. We have used pedestrian modelling to determine travel times, queuing levels, and Level of Service (people density) for a range of scenarios.

Below, our team examines two potential modelling approaches to new operational strategies for transit operators to predict whether social distancing can be reasonably practiced, and how risk might be mitigated.

Sample Transit Station

Managing people movement

First, we wanted to understand if the solution is as simple as telling passengers to maintain a two-metre separation. Unfortunately, the reality is that most existing infrastructure is not designed for such a low density of people while still operating at a pre-coronavirus pandemic capacity. This means we need to be more proactive. An example of this can be queuing on a train platform as passengers line up for stairs.

In pre-COVID operations, after a train arrives, large queues formed at the base of stairs while the rest of the platform remained empty – a scenario now considered high risk. In order to mitigate this, passenger volume will need to be dispersed in a manner that decreases this density.

Potential Solution One: Commuting, two metres apart

In the video above, two platforms are shown with trains arriving at identical times and with identical passenger volumes. Both trains are assumed to only be at 30% capacity, which roughly equals no standing passengers and only 50% of the seats occupied. All stairs and platform widths are identical. Platform B has not been assigned any mitigation strategies. Experience tells us that passengers all depart their train and queue at the base of the closest stairs.

Alternatively, Platform A has been assigned specific queuing behaviour. Passengers leave their train and begin to form queues along the platform at the base of stairs. This could be achieved with passenger education, directional signage, and enforcement by transit agencies. Throughout the simulation, passengers who come within two metres of another passenger turn red to indicate this proximity, and hence increased risk of exposure. Platform A with the queuing mitigation strategy has fewer red passengers, indicating this additional risk when compared to Platform B.

Operational impacts of social distancing

The low risk scenario presents several challenges. On Platform B, trains are empty, and all passengers are queuing, while trains on Platform A still have passengers waiting for space to safely move across the platform.

Comparison of time to clear the platform between two options 

A queuing strategy will dictate that trains would take longer to clear, and the platform would be occupied for a longer period. This will inevitably impact train schedules and, as a result, overall commute time for travelers.

Restrictions to overall vehicle capacity must also be taken into consideration. To ensure adequate space for passengers, transit agencies are already acting to lower capacity by blocking every other seat on trains. To reach full overall volume of commuters using lower capacity vehicles, shorter headway between arrivals and departures will be required. This is at odds with any mitigation strategies that increase the amount of time provided to clear a given space. This demonstrates the importance of understanding the interconnectedness of any mitigation strategy and quantifying those impacts as accurately as possible beforehand.

Potential Solution Two: Circular people flows

Another potential solution is to design circular people flows rather than linear flows, with the goal of minimizing people walking in opposite directions and crossing paths. In this scenario, stairs and escalators help to enforce a one-directional flow, and signage for horizontal elements (doors, corridors, and platforms) can communicate suggested walking paths. This strategy was also tested analytically.

In the video above, counterflow is introduced to the model where passengers are trying to board the same train that just arrived. A train arrives on both Platform A and Platform B at the same time, with identical passenger volumes both departing from and boarding each train.

On Platform B, all stairs are two-way, and no mitigation strategies have been added.

On Platform A, the left stair is one-way (down) and the right stair is one-directional (up). As well, passengers boarding the train are provided waiting space at platform level until each car is clear to board. This has promoted a circular passenger flow. In this model, a colour gradient has been added where passengers turn yellow when they are within 2.5 metres of another passenger and this transitions to orange and eventually red as that distance decreases. This can illustrate varying levels of risk.

Operational impacts of circular people flows

Implications of this strategy include a higher demand on stairs. The station was likely not designed to account for stairs being one-directional only. This may result in more queuing at the base of some stairs – a challenge in itself that would require a solution.

Passengers could also be provided with waiting space in concourses or mezzanines prior to going to a train platform while passengers who have just arrived clear the trains, but this too would require further study to understand whether the available space could account for all other passenger flows.

What’s next?

While the above has demonstrated possible strategies for transit infrastructure, each building or station will have its own unique challenges based on the existing geometry, ridership, operational strategies available to management, and the behavioural characteristics of the users.

In the coming weeks, we will be examining how Pedestrian Modelling can support remobilization strategies for other applications, like airports, institutions, commercial buildings, and sports and cultural venues. As a valuable tool for mitigating risk, we can test each potential strategy to understand the operational impact while protecting the well-being of our communities.

To learn more, contact Matt Smith.