A GA-SLP-Based Dynamic Allocation Method for Electric Power Emergency Materials Considering Disaster Impact Differences

Abstract

To address the challenge of supply–demand mismatch in emergency power materials following disasters and to improve the fairness of resource allocation, this study proposes a dynamic allocation method that integrates a Genetic Algorithm (GA) and System Layout Planning (SLP) while accounting for disaster impact differences. An evaluation system incorporating population distribution, disaster intensity, facility vulnerability, and material shortage levels is established. The CRITIC weighting method is employed to quantify the urgency of material needs at each disaster-affected site. A multi-objective optimization model is formulated to minimize the maximum shortage degree, prioritize high-urgency nodes, and reduce cross-regional transportation costs. The SLP method is used to generate an initial spatially feasible allocation scheme, and an improved GA is applied to perform a global search for Pareto solution sets and optimize path coordination. Experimental results show that the proposed method significantly enhances both the material guarantee rate and distribution fairness, particularly in high-urgency areas. The average hypervolume index of GA-SLP reaches 7.5642, which is notably higher than those of comparative algorithms. This approach effectively improves the scientific rigor and operational efficiency of power emergency material allocation.