It is very important to make a scientific optimum decision on the repairable spare parts inventory allocation for complex equipment, and seek the best balance between spare parts cost and availability in maintenance. Thus, the focus of this paper is on the analysis of multi-echelon spare parts inventory system, the goal of it is to build a multi-echelon spare parts inventory model, evaluate the related parameters and inventory index, and develop ant colony optimization algorithm (ACO) in order to identify stocking policies to minimize system-wide spare parts support cost subject to the average waiting time per demanded part at each of the locations. The experiment gives insights into the relative improvement achievement achieved by applying lateral transshipments.