A multiagent system based safe fire emergency evacuation guide: (a simulation for a modern office building in Kenya)
Evacuation of people in the event of hazard is one of basic problems of human society. In emergency scenarios there is limited time to act and or react; this results to extensive life loss because the time needed for safe evacuation in a threatened building was not available, or people running to the wrong direction of safety. In pursuit for a solution, systems have been developed to alert or warn occupants on the presence of a fire emergency so that they can act on time. Despite having such systems in place, efficiency in evacuation has not been realized, a fact that has opened room for more research and study. Studies indicate that a substantial number of deaths occur due to wrong decisions occupants make within the available evacuation time. Besides, they reveal that guiding occupants during evacuation proves to be more effective because it decreases the average escape time thereby increasing the chance of survival in fire emergency situation (Fahy & Proulx 2009). In Kenya, building management and building owners use the evacuation plans designed to them to aid evacuation during fire emergencies. In evacuation the plans are limited in that they do not explicitly tell you how but what to do. We propose to do a simulation of safe emergency evacuation guide (EvacSim) with the intention of determining evacuation efficiency. The simulation intends to highlight building fire disaster and how to achieve efficiency in evacuation. Efficiency translates to more lives saved within a short period of time. At extreme cases were available exits are blocked the simulation will show a location of safety as occupants await external assistance. EvacSim can be furthered to produce a system for similar purpose; Full adoption of the system will realize a good number of people being evacuated within a short time; As a result, there will be an increase in the number of evacuees safely evacuated while a decrease in the number of fatalities.