Emergency path planning based on improved ant colony algorithm

With the increasing complexity of building structures, scientific and effective emergency evacuation path planning plays a vital role in reducing casualties and property losses. Owing to the characteristics of positive feedback, strong robustness and low computational complexity, etc., Ant Colony Optimization (ACO) is very suitable for addressing emergency evacuation path optimization in complex buildings. However, the classical ACO has some limitations such as slow astringency and earlier stagnation, etc. Thus, a novel Emergency Path-planning Improved Ant Colony Optimization (EPIACO) algorithm is proposed. EPIACO contains six improved mechanisms including uneven initial pheromone distribution, heuristic function with directional judgment, adaptive pheromone volatility factor, differentiated pheromone update rule, improved state transition probability strategy, and path smooth. Subsequently, a series of comparative simulation experiments between EPIACO and eleven existing algorithms are conducted in different grid maps to test and verify the performance of EPIACO. The experimental results demonstrate the advantages of EPIACO in terms of decreasing the path length, reducing the number of turning times, and accelerating the convergence speed. Finally, a user interface is developed to facilitate the human-machine interaction and demonstrated with a fire evacuation case. The case study highlights the practicability and flexibility of the user interface and the proposed algorithm in the process of dynamic evacuation path planning considering fire spread and personnel density change. This work highlights a feasible emergency path planning method, whose characteristics of fast response and dynamic optimization can well promote the development and application of emergency expert systems in real evacuation scenarios of buildings.