A model based estimator for inventory tracking with shrinkage detection in perishable supply chain

Accurate inventory tracking in the presence of shrinkage is a fundamental prerequisite for an efficient management of perishable Supply Chains (SC)s. In this paper we propose an effective low-cost alternative to the widespread use of Radio-Frequency IDentification (RFID) technology. The new presented method is based on the series connection of a physical low cost sensor with an estimator and on an outlier detector. Whenever the outlier reveals the presence of shrinkage, the internal state of the estimator is reinitialized to improve convergence towards the new value of the inventory level. We test our approach on a perishable SC governed by an Order-Up-To (OUT) policy