Agent-based models: How can they reduce the risk posed to people by extreme flood events?

Tagg, A. and Lumbroso, D. and Davison, M. (2017) Agent-based models: How can they reduce the risk posed to people by extreme flood events? In: ICFM7, 5-7 September 2017, Leeds, UK.

Full text not available from this repository.


As the effects of major flood events continues to increase around the world, driven at least in part by climate change, the importance of considering a wide range of response measures has never been more pressing. Over the past 10 to 20 years there has been a paradigm shift from flood defence to flood risk management, and that use of non-structural measures can be more cost-effective, particularly when dealing with residual flood risk. One management response is to plan for evacuation of affected communities so as to reduce risk to people and with it loss of life. For this to be effective and safe, emergency plans need to be based on realistic assumptions of human behaviour, and to have been tested for a range of event scenarios. Agent-based modelling of human responses before and during flood events is a powerful tool for considering risk to people, and its use has grown in the past couple of decades. This paper describes the application of the Life Safety Model in many countries and for different flood types, including dam failure in Malaysia, river flooding in Australia and coastal surges in the eastern USA. The paper considers how the model attempts to capture the physical and behavioural rules that may apply during evacuation, and the uncertainty inherent in any modelling approach. The LSM software combines a widely-used transport module with physical and behavioural rules that are in part informed by limited observations. So the impact of floodwater on loss of life and injury, in terms of its depth and velocity, is derived from laboratory experiments that tested human stability over a range of conditions. Even so, this testing does not cover all age groups or physical characteristics, and so there remains uncertainty in the LSM simulations. Similarly, the model can specify the behaviour of people during major floods, such as time taken to start the evacuation and mode of evacuation. The model is therefore a powerful tool in being able to consider how such assumptions affect loss of life and evacuation times. But further research and experiential evidence would help in informing such behaviour to include in the model. The LSM is currently being used in several flood applications for emergency management planning, to assess risks to people and to try and impose a response on the affected communities to minimise fatalities and injuries. The paper will provide examples to illustrate this. Key issues are the assumptions on response times to the warnings, and the effect this has on the road usage. The case studies will illustrate how the location of warning centres and safe havens affects the loss of life. The LSM is an established and powerful agent-based tool that permits the simulation of behaviour before and during a flood, that can be used in emergency management. It uses standard physical rules on the stability of people, and allows a range of behavioural decisions to be tested.

Item Type: Conference or Workshop Item (Paper)
Additional Information: 7th International Conference on Flood Management (ICFM7)
Uncontrolled Keywords: Agent-based modelling; emergency management; evacuation; loss of life
Subjects: Floods > Flood incident management
Divisions: Floods
Depositing User: Unnamed user with email
Date Deposited: 02 Apr 2020 09:53
Last Modified: 29 May 2020 07:43

Actions (for site administrators only - login required)

View Item View Item