In this paper, the two-echelon Inventory Routing Problem（IRP） within the mode of Vendor Managed Inventory（VMI） is studied in accordance to the characteristics of demand fluctuation. A mixed-integer programming model on two-echelon IRP with heterogeneous fleet is developed to minimize the total system cost of a two-echelon IRP. An improved particle swarm optimization algorithm is designed to solve the model. Numerical experiments verify the applicability and effectiveness of the model and the algorithm proposed in this paper. The results indicate that using heterogeneous fleet can not only increase the loading rate of distribution vehicles and decrease the inventory levels in retailer warehouses, but also reduce the total system cost of the two-echelon IRP. Sensitivity analysis shows that no matter how the demand fluctuates, the total system cost of a two-echelon IRP will be reduced when using heterogeneous fleet.

%U http://cea.ceaj.org/EN/10.3778/j.issn.1002-8331.2005-0235