Diary of an Evacuee

February 20, 2019

People move in and around buildings during emergency and non-emergency scenarios. Whether it is walking to the check out in a supermarket or descending a staircase in response to a fire alarm, people move to achieve a goal of some sort. Emergency and non-emergency scenarios are not unrelated: they are instead highly-coupled.

The key connection between routine and emergency situations is you - the individual, your decision-making and your ability to respond. Normally, these abilities are focused on using the space in a routine way – getting a sandwich or sitting in the cinema. During an emergency these same abilities are used to reach a place of safety.

An evacuee during an emergency was only recently a pedestrian routinely using the same space. Picture yourself sitting at your desk at work or at a dinner table in a café when afire alarm sounds indicating that an incident has occurred – that the scenario has changed.  You move from the routine to the unexpected in a matter of moments. You bring with you the person you were before the alarm to your response to the alarm. To quantify the outcome of an evacuation it is critical that we understand pedestrian abilities, decision-making and performance during the lead up to it and during the emergency itself, and then represent this understanding in our analysis and calculations.

Understanding pedestrian performance – your performance – initially provides insights into the effectiveness of building design and management. This understanding allows us to enhance building design for routine use through estimating performance given different designs. For instance, understanding how you use the facilities (e.g. the bathroom) on the approach to an arena will help us better plan such facilities in the future and enhance your experience (e.g. how many bathrooms are required and where should they be located).  

Pedestrian performance during routine scenarios also affects the initial conditions of emergency scenarios. For instance, identifying how you might routinely use a building will help identify where you (and the rest of the population) might be and what you are doing when an alarm sounds. This allows us to use these conditions as a starting point and then estimate how conditions might evolve during an emergency; i.e. estimate the time it takes you to reach a place of safety. It is difficult to estimate the spread of an occupant population throughout a building from the design alone. The location of facilities, services and access will influence how you use the building – deviating from these design expectations. For instance, assuming a population is spread evenly across an airport concourse presents a very different situation from the population instead queuing at the coffee shop and the bathrooms located at one end of the concourse, should an alarm sound.  

Therefore, the ability to establish how people routinely use a space (and how a population is distributed about a space) has enormous benefits both for pedestrian management (to enhance performance and pedestrian experience) and evacuation planning (as an indication of initial spread of people about a building when the alarm sounds). Understanding individual decision-making will also provide insights into what evacuees do once the alarms sounds; e.g. how long it takes them to respond to the alarm, what routes they use when they move off and how quickly they might move. This is fundamental to identifying the time for a population to reach a place of safety using an engineering calculation or a computer simulation. Understanding pedestrian decision-making has practical value in enhancing building design for routine use and enhancing safety during emergency scenarios, by informing our performance assessment. By allowing us to better account for the response of the population under different circumstances.

Recent research undertaken by Movement Strategies has involved collecting data on routine movement and developing models for evacuation response. We recently worked with Lund University students to derive the speeds at which visitors moved around a UK stadium across several different events. These speeds were influenced by the number of people present and the type of population present. This research informs our understanding of pedestrian performance and supports the development of tools to calculate the consequences of such performance on the conditions experience; e.g. the time spent queuing.

We have also been working with various partners (NFPA, Imperial College, etc.) to develop a modelling platform for quantifying large-scale evacuation from wildfires. This is requiring a review of existing data and modelling tools. Part of this is developing a response model of people’s decision-making process when confronted by a fire; i.e. estimating the time for a resident to mentally move from a routine to an emergency scenario and initiate a response.  

It is important to continually update our understanding – conducting research and learning from the research of others – and to ensure that the tools we use to estimate performance are as credible as possible. And, of course, to ensure that we better understand how pedestrians perform under different scenarios and how we can take this into account in our work – to enhance routine experiences and enhance safety during less routine situations.