Sensing, calculating, and disseminating evacuating routes during an indoor fire using a sensor and diffusion network

We present a framework for calculating efficient routes to evacuate a single-story building during a fire. The presented approach includes the processing of information collected from a sensor network, deployed with high granularity in an indoor environment, an algorithm to calculate evacuation routes, and dissemination of route information to occupants. The deployed sensors determine the state of whether a well-defined area of a building is transitable or blocked. The collection of transitable areas is used to determine evacuation paths. Route calculation is based on the A* algorithm. We evaluate the performance of the proposed framework under the event of a fire. We consider routing with safety margins on the distance between an occupant and the fire to increase the probability of success of evacuation under a spreading fire. We also show that the information provided by a fully functional information network adopting the proposed framework facilitates the evacuation to the extent that an occupant experiences little or no confusion as to where to exit for fires spreading at different speeds. In general, we see that an occupant may easily be routed to the exit to which evacuation may take the shortest time and that changes of evacuation routes hardly occur.