Abstract: To optimize emergency logistics distribution system for natural disasters like floods and earthquakes, a bi-level programming model of location allocation-vehicle routing problem is set up by considering the characteristics that government and enterprises participate together but make decision separately. The upper level is to minimize the total time of emergency logistics system, and the lower level is to minimize the sum of distribution costs and time penalty cost. According to the characteristic of independent decision-making and interaction effect in upper and lower level in the model, the problem is broken up into two stages. In the first stage, viable gathers of distribution center are produced. In the second stage, vehicles and transport routes are arranged under viable gathers. A heuristic rule of two stage hybrid simulated annealing algorithm, one-phase tabu search,is designed to determine feasible emergency distribution center collections. The initial solution is construeted by principle of greedy proximity. Code is exchanged to search neighborhood feasible solution. The optimal solution is recorded and updated unloading and driving time is accumulated and output as the basis of the upper decision. Finally a numerical example and the comparing hybrid genetic algorithm are given to verify the validity of the model and the feasibility of the algorithm. A solution for emergency logistics system planning problem of multi-level decision is provided.

Key words: emergency logistics system, location allocation-vehicle routing, bi-level programming, tabu search, simulated annealing algorithm