This paper focuses on the optimization of the picking process in a real refrigerated warehouse of a food wholesaler operating in the UK. The warehouse layout is unusual, as the warehouse has only one cross-aisle at the front of the shelves and no cross-aisles at the end of the rack, which, instead, ends in a blind wall. Because of this layout, traditional routing policies (e.g. the Sshape or largest gap policies) are not applicable. The current policy adopted for order preparation is batch picking & sorting. By means of an ad hoc simulation model, developed in Microsoft ExcelTM, we reproduced the warehouse layout and tested 48 alternative picking configurations, with the aim of enhancing the performance of the current order preparation process. Because routing cannot the varied so much, the analysis carried out focuses on the allocation of product in the warehouse, the use of classes for grouping products, and on an alternative picking policy, called “area batch picking”. From the results obtained through simulation, configuration deserves particular attention for a practical application, as it allows the total picking time to be decreased by approx. 223 minutes per day.