Smoke generated by fires in buildings frequently creates a greater threat for occupants than does the heat. Reduced visibility and eye irritation caused by smoke makes it difficult for an occupant in a building to locate escape routes and exits. Therefore, studies of predicting smoke movement inside a building during a fire has become very important. An approach to doing this is to develop a hybrid computer-based model to effectively predict the movement of smoke which combines the accuracy of a zone based model near the fire and the efficiency of a network based model sufficiently far away.

The objective of this research is to develop a hybrid of zone and network fire modeling which would be able to simulate the smoke movement from fires in multi-compartmented buildings.