Application of multi-objective artificial gorilla troops optimization algorithm in intelligent evacuation path planning for building fires

Abstract

In the event of a sudden incident in large buildings, optimizing personnel evacuation paths has a direct and significant impact on overall evacuation efficiency. To address personnel evacuation scenarios caused by building fires in complex environments, this study presents a path planning approach based on the Multi-Objective Gorilla Optimization (MIGTO) algorithm. The objective function comprises path risk, congestion cost, and path length. However, traditional GTO algorithms face limitations in addressing such problems, including slow convergence, premature convergence, and susceptibility to local optima. To address the aforementioned challenges, this study introduces variable updates to the spiral position during the algorithm development phase to improve search efficiency. The nonlinear parameters are adjusted to achieve a harmonious balance between global and local search capabilities. In addition, the alert perturbation mechanism is implemented before each optimization iteration, which improves the algorithm's ability to overcome local optima. Furthermore, a compound chaotic mapping is introduced to increase the diversity of the population. Experimental results demonstrate the outstanding performance of the MIGTO algorithm in various grid map path planning scenarios. Whether faced with complex obstacles, multiple exits, or multiple fire sources, the path inferiority indices of MIGTO in path planning are significantly superior to other algorithms. The MIGTO algorithm is not only suitable for dynamic planning of fire evacuation paths, but can also be used in the Pathfinder evacuation simulation software, where it outperforms the software's built-in algorithms.