An enhanced cycle counting approach utilising historical inventory data

Inventory record inaccuracies are an issue in both retail and warehousing and can lead to considerable costs due to stock-outs and excess inventory. In order to ensure that inventory record inaccuracies are minimised, warehousing organisations conduct inventory checks on items in the warehouse to identify and then rectify misalignments between the data records and the actual physical items. The challenge is that inventory checks are extremely time-consuming activities, especially when all items need to be checked. In order to reduce this effort, cycle counting methods can be employed which check only a sample of the items. In this paper we propose a new cycle counting method which aims to identify inaccurate items using historical inventory checking data. The method aims to discover correlations between various properties of the items (e.g. storage location, frequency of movement etc.) and inaccuracies. These correlations can then be used to identify which items are most likely to be inaccurate so inventory checking can focus on those items. The approach is evaluated based on actual data from warehousing companies.