A study of herding behaviour in exit choice during emergencies based on random utility theory

Modelling human behaviour in emergencies has become an important issue in safety engineering. Good behavioural models can help increase the safety of transportation systems and buildings in extreme situations like fires or terrorist attacks. Although it is well known that the interaction with other decision makers affects human behaviour, the role of social influences during evacuations still needs to be investigated. This paper contributes to fill this gap by analysing the occurrence of Herding Behaviour (HB) in exit choice. Theoretical explanations of HB are presented together with some modelling approaches used in different fields where HB is relevant. A discrete choice stated preference experiment is then carried out to study the role of HB in the decision-making process concerning exit choice during evacuation. A binary logit model is proposed showing that the occurrences of HB are affected by both environmental and personal factors. In particular, the model shows that the personal aptitude to HB can have a key role in selecting an exit.