Optimizing Emergency Response and Evacuation Planning Using Swarm Intelligence and Agent-Based Modeling in High-Risk Industrial Facilities

Background: High-risk industrial environments, such as the Esfahan Steel Company (ESCO), face significant challenges regarding emergency management due to complex layouts and the presence of hazardous materials. Traditional evacuation protocols often overlook the stochastic nature of human behavior and real-time environmental dynamics, leading to suboptimal safety outcomes during critical incidents like gas leaks or structural fires. Objectives: This research aims to develop a sophisticated framework for optimizing emergency response and evacuation planning within the specific context of the steel industry. The primary goal is to minimize total evacuation time and prevent fatal bottlenecks in high-density work zones. Methodology: The study proposes a hybrid computational approach integrating Agent-Based Modeling (ABM) and Swarm Intelligence (SI). Using ABM, individual workers are modeled as autonomous agents with heterogeneous characteristics, including varying stress levels, movement speeds, and environmental familiarity. Simultaneously, a Particle Swarm Optimization (PSO) algorithm is utilized to dynamically calculate the safest and most efficient exit routes. The simulation is calibrated using the physical spatial data of ESCO’s primary production units to ensure industrial relevance. Results: Preliminary simulation results indicate that the integration of swarm-based pathfinding significantly outperforms conventional static signage methods. The hybrid AI model demonstrates a reduction in total evacuation time by approximately ۱۸-۲۵%. Furthermore, the model successfully identifies "critical congestion points" within the facility, allowing for the strategic redesign of safety corridors and assembly points. Conclusion: By leveraging the predictive power of Artificial Intelligence, this study provides a robust decision-support tool for HSE managers. The findings suggest that AI-driven evacuation planning not only enhances worker safety but also strengthens the overall disaster resilience of large-scale metallurgical complexes.

کلیدواژه ها:

Emergency Evacuation ، Swarm Intelligence ، Agent-Based Modeling (ABM) ، Esfahan Steel Company ، Industrial HSE ، Risk Mitigation