A Dynamic Graph Model-Based Method for Special Equipment Resource Scheduling

Abstract

As a critical component for ensuring safety in major events, special equipment resource scheduling faces challenges like dynamic status changes, environmental disturbances, and multi-entity collaboration, making traditional static methods insufficient for real-time response. This paper proposes a dynamic graph model-based method, constructing a multi-layer directed weighted graph with task, environment, and resource nodes to characterize their couplings. An adaptive weight update mechanism optimizes paths in real time by integrating risk levels, environmental intensities, and resource availability. A multi-objective model with response time, utilization, and risk is built, solved by particle swarm optimization. Experiments in dynamic scenarios show 12%-17% higher resource matching rates and 13%-20% shorter durations than traditional methods, offering a scalable framework for safe scheduling.