A Multi-Objective Robust Optimization Design for Grid Emergency Goods Distribution Under Mixed Uncertainty

Emergency goods distribution plays an important role in the grid emergency relief command system. However, the traditional experience-based distribution plan currently in the power grid cannot meet the increasing demand for the types and quantities of emergency goods, meanwhile, most studies ignore the uncertainty in distribution parameters and diversification of distribution objectives, which causes a gap between theoretical research and practical application. Therefore, this paper establishes a model to guide the logistics design for transferring relief supplies. The model first assesses the importance of affected areas for determining distribution priority based on the electrical characteristics. To better simulate reality, the uncertainties in demand, supply, and the costs of procurement and transportation are considered. Additionally, the model features three objectives: shortening the travel time, reducing the goods shortage, and saving the total cost of pre- and post-disaster phases. In order to handle the uncertainties, the robust optimization approach is utilized. The numerical example is solved with ε-constraint exact method, and this case illustrates the specific process of goods distribution, the relationship between objectives, and the sensitivity analysis of uncertainties. For the large-size forms, two heuristic algorithms are proposed and the efficiency of the proposed algorithms is assessed.