Time-critical decentralised situational awareness in emergencies: an adversarial biosecurity scenario

Abstract

Abstract Crises in a global setting of interdependencies call for time-critical coordinated responses. However, it is often the case that the mechanisms responsible for these actions do not agree across all their hierarchies. This can be roughly attributed to personal estimations of the situation and to social influence. An ensuing lack of consensus against crises can be dire and echo across entire populations. One such instance is the case of biosecurity threats. A particularly interesting class of threats lie within urban environments, which tend to fall within the scope of bad actors. With this work we aim to computationally contribute to the understanding of the dynamics of perceived danger formation among agents responsible for responding to ongoing biological attacks in urban settings. We assume this perception is a function of a personal estimation of local information about the danger and of social influence stemming from the agents in question framed in an agent-based model. The simulations point towards a high dependence of perceived dangers on the personal estimations of the agents. The conditions under which the perceived dangers deviate from the real ones are explored over a range of assumptions on personal measurements and several dispositions towards the influencing environment. The insight provided by these results at the individual and collective level set the tone for further investigation on such behavioural phenomena, providing a flexible computational framework addressing generic threats (true dangers) in a time-critical context.

Keywords

• Agent-based modelling
• Danger perception
• Estimation bias
• Social influence
• Urban environments