Abstract: A bi-level distribution network model of multiple products under uncertain demands is proposed for minimizing distribution cost, inventory cost, shortage cost, and penalty cost. With actual situation taken into consideration, factorys production capacity constraints, minimum and maximum transfer quantity constraints, minimum distribution quantity constraints are considered. Due to the complexity of the model, a tabu search algorithm is proposed to solve the model. The correctness of the model is verified by a numerical example. By changing the regret value coefficient, the best robust solution is found for the numerical example. It takes less than 1.5 seconds to do so, which shows the efficiency of the proposed algorithm. This is significant for practical application.

Key words: distribution network, locationallocation, uncertainty, optimization, tabu search