A new heuristic algorithm to improve the design of a vertical storage system

In this paper, a new heuristic algorithm to support the design phase of a vertical storage system is presented. The considered vertical storage system is made up of racks accommodating metal containers which are arranged on the two sides of the lift device that feeds them. The proposed heuristic algorithm has been developed combining two well-known problems: Bins Packing Problem (BPP) and Rectangular Nesting Problem (RNP). Given a list of products that must be stored, the dimensions of the warehouse's racks and its load capacity, the developed algorithm allows to obtain the list of products that must be placed in each stock keeping unit and returns the right position and orientation any item should take. The algorithm provides the minimum number of racks with height dimension as low as possible. This aim is due to the common interest of automated vertical storage systems designers or owners in lowering costs, which trend to grow up with racks number and height. Furthermore, the position of the items inside each rack is managed to optimize the volume exploitation and to balance the container distributing the weight inside it. The whole procedure also regards the maximum weight constrain that basically limits the filling of loading units. The robustness of the proposed algorithm has been studied simulating different scenarios, by changing boundary conditions such as the number of items to be stored, their middleweight, their average size and the variance of these physical characteristics. Finally, the algorithm has been applied in a realistic situation to support automated vertical storage system design aimed at holding in stock metal moulds.