Integrated scattered storage and picker routing in picker-to-parts warehouses

Abstract

Scattered storage is a common operational approach in business-to-consumer online retailing, where each Stock Keeping Unit (SKU) is stored at multiple locations within the warehouse. This approach aims to increase the likelihood of having items per SKU nearby consistently, thereby reducing unproductive walking time during order picking. However, reaping these benefits necessitates additional decision-making in selecting items while planning the picker’s route. In addition, previous studies have tackled storage assignment and picker routing independently. In this paper, we investigate the integrated scattered storage, picking selection and picker routing problem and three special cases with predefined routing policies (return, S-shape, and midpoint) in picker-to-parts warehouses. The objective is to minimize the weighted sum of travel distance for both storage and order picking. These problems are formulated as mixed-integer programming models and proved to be NP-hard. To solve real-size instances, we propose an efficient adaptive large neighborhood search algorithm, where the commercial solver is used in the repair step, so that the new solution after the repair step is always feasible and never worse than the current one, and the search procedure becomes faster. Through numerical experiments, it is demonstrated that the algorithm yields high-quality solutions in a reasonable computation time. Compared to methods commonly used in practice, our algorithm can reduce the objective value by up to 14%. Moreover, our findings underscore the potential value of a simultaneous approach for enhancing the efficiency of operations in warehouses.