Joint replenishment problem (JRP) is a type of inventory model which aims to minimize the total inventory cost consisting of major ordering cost, minor ordering cost and inventory holding cost. Different from previous papers, this study considers one warehouse, multi items and N retailers. An innovative heuristic approach is developed to solve the problem. In this paper, we consider a multi echelon inventory system and seek to find a balance between the order cost and the inventory holding costs at each installation. The computational results show that the innovative heuristic provides a near exact optimal solution, but is more efficient in terms of the computational time and the iteration number.