NetLogo implementation for Crowd Evacuation

Issues related to crowd evacuations have become extremely important and gained the attention both of practitioners and public authorities. Understanding the dynamics and the emerging behaviour of the crowd during evacuation is a challenge under any circumstances. By creating a synthetic population it is possible to employ agent-based modelling (ABM) techniques to build artificial laboratories [1] for experimentation in-vitro. ABM relies on the modelling of micro level entities and dynamics. The observed global behaviour of the system emerges as the result of these micro level dynamics. Such an approach represents a unique way of testing theories and models of crowd behaviour in various ”what-if” scenarios. In their common environment pedestrians tend to show some basic attributes.The basic principle is the ”least effort principle”, which means everyone tries to reach their goal as fast as possible spending the least amount of energy and time. Observations made for crowds in emergency situations feature typically the same patterns [2]. Within this context, an agent-based implementation of a crowd evacuation scenario using NetLogo [3], a platform that permits rapidly the creation of simulation prototypes, will be discussed. Although the model may appear to be simplistic, its conceptualisation encompasses many aspects of the observed system in the real world. As such, we are able to observe typical emerging behaviour patterns during evacuation situations, namely arching, clogging and other herding phenomena Our aim is to show that models developed using NetLogo, albeit simple can be expanded and adapted for fire safety experts to test various scenarios and validate the outcome of their designs.