Inventory early warning agent in a supply chain management system
AssigneeSap SE
InventorsAlexander Renz, Ye Chen
Life of this patent
Estimating the term…
Abstract
An inventory agent operating on software includes instructions operable to cause a programmable processor to receive inventory data relating to stock in an inventory, apply the inventory data to a conditional probabilistic predictive statistical algorithm, calculate a predicted inventory level, and use the calculated predicted inventory level to determine whether to order additional stock for the inventory. The statistical algorithm uses a conditional probabilistic model to process the data. The inventory agent may be implemented in a supply chain management system.
Claims39
A computer-implemented method comprising:
- applying machine learning techniques to historical supply chain data to build a conditional probabilistic model including patterns of behavior related to consumption and replenishment activities in a supply chain;
- receiving inventory data relating to stock in an inventory;
- applying the inventory data to a conditional probabilistic predictive statistical algorithm, wherein the conditional probabilistic predictive statistical algorithm uses the conditional probabilistic model to process the inventory data and to calculate a predicted inventory level based on a supply prediction and a demand prediction, the supply prediction factoring in variability in supply chain activities;
- using the calculated predicted inventory level to determine whether to order additional stock for the inventory;
- and
- automatically ordering a replenishment of the inventory when the calculated predicted inventory level falls below a predetermined minimum.
A system comprising one or more computer systems and an inventory agent computer coupled to the computer systems over a network, the inventory agent computer being operable to:
- apply machine learning techniques to historical supply chain data to build a conditional probabilistic model including patterns of behavior related to consumption and replenishment activities in a supply chain;
- receive inventory data relating to stock in an inventory;
- apply the inventory data to a conditional probabilistic predictive statistical algorithm, wherein the conditional probabilistic predictive statistical algorithm uses the conditional probabilistic model to process the inventory data and to calculate a predicted inventory level, an upside 10% confidence bound of the predicted inventory level and a downside 10% confidence bound of the predicted inventory level, based on a supply prediction and a demand prediction, the supply prediction factoring in variability in supply chain activities;
- and
- use the calculated predicted inventory level to determine whether to order additional stock for the inventory.
A machine readable medium having instructions therein which when executed by a computer cause the computer to perform a set of operations comprising:
- applying machine learning techniques to historical supply chain data to build a conditional probabilistic model including patterns of behavior related to consumption and replenishment activities in a supply chain;
- receiving inventory data relating to stock in an inventory;
- applying the inventory data to a conditional probabilistic predictive statistical algorithm, wherein the conditional probabilistic predictive statistical algorithm uses the conditional probabilistic model to process the inventory data and to calculate a predicted inventory level based on a supply prediction and a demand prediction, the supply prediction factoring in variability in supply chain activities;
- using the calculated predicted inventory level to determine whether to order additional stock for the inventory;
- and
- automatically ordering a replenishment of the inventory when the calculated predicted inventory level indicates a likely undesirable variation in inventory.
Dependencies are inferred from the claim text.