Simulating Evacuation of Humanitarian Settlements

Informal settlements and settlements of displaced communities (e.g., humanitarian settlements) provide crucial shelter for people, including those negatively impacted by natural hazards and human-caused crises. However, these settlements are also prone to fire as a ‘secondary incident’. This occurrence can be influenced by the use of flammable materials in the building construction and unsafe fire and electricity practices given the harsh and sudden conditions faced (along with pre-existing economic hardship). Humanitarian practitioners and agencies are increasingly recognizing the importance of addressing the fire safety problem in humanitarian settlements, with a particular focus on community evacuation. This article presents a new method to assess the safety of the areas involved and explores the use of a pedestrian simulation model (Pathfinder), to conduct evacuation analysis in an example settlement given the occurrence of a fire. The goal is to demonstrate the potential for such applications and provide a foundation from which such an application might be formalized and tested across representative scenarios. The Pathfinder tool is widely used in the fire safety community for building evacuations. It is applied here on a larger scale, although examining the same core evacuation factors. The impact of four key factors on evacuation performance is examined (i.e., density, pre-travel delays, route choice, and restricted routes), based on a case study of the community within the Kutupalong refugee camp in Cox’s Bazar, Bangladesh. Study results show that Pathfinder provides insights into evacuation outcomes during settlement fires despite the complexity involved in creating the simulation model. The results also show that the evacuation times produced were sensitive to the four conditions tested and that the prolonged evacuation times resulting from these conditions could lead to serious consequences to settlement residents, especially in cases with fast moving fires. Of course, the accuracy of such estimates is reliant on the data available and the assumptions made to configure the model. However, we have demonstrated that the simulation tool can cope with the scenarios examined and provide insights into the evacuation dynamics produced—establishing the potential of such a tool and the value of more rigorous examination.