Routing and storage assignment for the precedence-constrained order picking process

Abstract

Order picking is retrieving items from the warehouse to fulfill customers’ orders. It is considered the most labor-intensive and time-consuming operation in a warehouse and composes almost half of the total operating costs. Thus, developing efficient routing sequences for order pickers has been one of the main focus projects of managers. In addition, in real-warehouse environments, routing is frequently influenced by precedence constraints. Precedence constraints arise when certain products need to be collected before others due to a particular physical characteristic of the items. For instance, precedence constraints may be defined by the products’ fragility, weight, or size, among others. Even though many warehouses face such constraints, they have often been neglected in the scientific literature. This dissertation is inspired by a practical case of a Mexican Company that stores perishable products, which are considered sensitive items; this means that they are easily deformed if a certain weight is placed on them. This situation arises the problem that the warehouse under study must consider Unit of Measurement and Load constraints. The Unit of Measurement constraint prevents box-packed items from being placed on top of individual units. Load constraint allows only a limited number of boxes to be placed on top of another box. To develop a solution to this concern, we propose a mathematical model to formulate the problem. Due to its complexity, the implementation of an approximate method was mandatory. Indeed, a Genetic Algorithm was designed to meet this problem’s requirements. In addition, we propose three Storage Assignment strategies to analyze if these further improve the traveling distance of the resulting routing sequences. These were applied to a set of instances obtained from the Company’s Warehouse Management System observations. We assess the picker routing and storage assignment strategies’ performance and obtain essential knowledge for this type of problem.