Multiobjective evacuation route assignment model based on genetic algorithm

Emergency evacuation in public places becomes a research focus in recent decades. One major objective of evacuation is to maximize the efficiency of the whole evacuation system. The paper proposes a multiobjective evacuation route assignment model to plan an optimal egress route set for the individual evacuees. The three objectives in the proposed model are to minimize the total evacuation time, minimize the total travel distance of all the evacuees and minimize the congestion during the evacuation process. These objectives need to be satisfied simultaneously while some of them conflict with each other. The travel speed on each road segment is related with the time and the number of evacuees on it during a certain time period. The congestion of a road segment is modeled as the density of evacuees passing it in time and space dimensions. The evacuation route assignment problem can be treated as a combinatorial optimization problem. A multiobjective optimization model based genetic algorithm is adopted to solve the proposed evacuation routing problem. Wuhan Sport Center in Wuhan city of China was taken as the experiment scenario to test the performance of the proposed algorithm. The results showed that it can provide some system optimal evacuation plans. Meanwhile, the multi-objective optimization model based genetic algorithm can produce a pareto optimal set rather than single optimal point, thus the model can give alternative strategies for the evacuation policy makers.