Optimizing evacuation path planning in high-rise building fires using an improved ant colony algorithm and dynamic window approach under smoke control scenarios

Traditional emergency evacuation strategies are constrained by limited adaptability to dynamic conditions and inefficient allocation of rescue resources. To assess the impact of smoke control facilities during high-rise fire evacuations, Building EXODUS software was used to simulate occupant behavior under the influence of air curtains and mechanical smoke exhaust systems. Furthermore, to address real-time obstacle avoidance in dynamic environments and during group movement in fire emergencies, a multi-person evacuation path planning method was developed by integrating an improved ant colony optimization algorithm with the dynamic window approach. This method incorporates a heuristic function that considers fire conditions and potential fields, along with a dynamic pheromone update strategy featuring reward and punishment mechanisms. The results indicate that the effectiveness of smoke control facilities in improving evacuation outcomes ranks as follows: combined operation of mechanical smoke extraction and air curtains, mechanical smoke extraction alone, and air curtains alone. Compared to the standard ant colony algorithm, the improved algorithm reduced planned path length by 7.71 %, decreased turn times by 73.30 %, and improved computational efficiency by 6.56 %. Furthermore, under scenarios with and without operational smoke control facilities, the improved algorithm combined with the dynamic window approach reduced path overlap areas by 11.76 % and 18.19 %, respectively, and shortened path lengths by 13.49 % and 14.19 %, respectively, compared to the Building EXODUS. The proposed method effectively addresses dynamic fire environments in high-rise buildings and provides a theoretical foundation for intelligent occupant evacuation.