Font Size: 

In the era of Industrial Revolution 4.0, the Internet of Things (IoT) play an important role in developing smart city. There are many kinds of IoT sensors that can help us in measuring some physical quantities, e.g. heat and smoke detectors. They are used as an input for early fire alert in a building, and furthermore, it is used in the analysis of optimal handling fire evacuation. In this paper, we utilize these sensors for building a Smart Fire Evacuation Management System (Smart-FEMS) by using swarm-based multi-agent simulation. We also conducted some evacuation experiments on public school buildings in Jakarta, Indonesia, which have a standard layout, such as classroom and corridor, to check the validity of simulation results with additional information about the occupant profile (teachers and/or students). The experimental results show that the Smart-FEMS scenario has the lowest number of dummy fire casualties. However, the actual number of dummy fire casualties is slightly different with Smart-FEMS results. We then examined the assumptions used in the simulation and, finally, we need to determine the appropriate parameter values for each condition.