EVACUATION ROUTE PLANNING MODEL BASED ON THE FUZZY ANT COLONY OPTIMIZATION ALGORITHM

Abstract

The construction of evacuation routes from urban areas has become an urgent task in the modern world, as the number of emergencies is constantly increasing. Standard route planning algorithms do not meet the requirements of fast and efficient evacuation because they do not fully consider the environmental parameters and have a high computational complexity, and erroneous results can have critical consequences, including loss of life. The paper proposes a model of finding an optimal evacuation route in emergencies in urban areas based on a modified ant colony optimization algorithm: an ant (a person or a vehicle) is allowed to start moving from several possible vertices of the graph, as well as to end the route at several available vertices. It is connected with the fact that there are usually multiple evacuation start points and destinations. The transitions matrix is built using additional parameters, the uncertainty of which is taken into account using the methods of computing with words and the theory of type-2 fuzzy sets and systems. For modeling, in the paper such additional parameters as the quality of the road surface, the number of road lanes, the level of traffic jams, and the distance to the epicenter of the emergency were used. The proposed model was implemented and applied in one of the quarter of Kyiv.