US 12,450,549 B1
Bayesian and frequentist anomaly detection ensemble
Phani Mitra Bulusu, Bangalore (IN); Rashid Puthiyapurayil, Bangalore (IN); and Vidhi Chugh, Delhi (IN)
Assigned to Blue Yonder Group, Inc., Scottsdale, AZ (US)
Filed by Blue Yonder Group, Inc., Scottsdale, AZ (US)
Filed on May 18, 2022, as Appl. No. 17/747,768.
Claims priority of provisional application 63/190,420, filed on May 19, 2021.
Int. Cl. G06Q 10/0833 (2023.01); G06N 7/01 (2023.01)
CPC G06Q 10/0833 (2013.01) [G06N 7/01 (2023.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method for applying machine learning to identify anomalous supply chain data, comprising:
scanning, by a networked imaging device comprising an imaging sensor, one or more items to generate a mapping of the one or more items in a supply chain, wherein the anomalous supply chain data results, at least in part, from the scanning using the imaging sensor;
generating, by a computer comprising a processor and memory, a probabilistic graphical model, based on supply chain states of the supply chain comprising one or more supply chain entities, to represent a performance of the one or more supply chain entities in the supply chain;
standardizing, by the computer, input features data related to the probabilistic graphical model;
performing, with the computer, data anomaly detection within the probabilistic graphical model using one or more frequentist data anomaly detection algorithms;
performing, with the computer, data anomaly detection within the probabilistic graphical model using one or more Bayesian data anomaly detection algorithms;
combining, with the computer and according to one or more weighting methods, the data anomaly detection performed using one or more frequentist data anomaly detection algorithms with the data anomaly detection performed using one or more Bayesian data anomaly detection algorithms;
detecting, with the computer and in response to the combining, an anomaly within the standardized input features data, wherein the anomaly is based, at least in part, on the scanning of the one or more items;
adjusting, with the computer, the standardized input features data in response to the detection of the anomaly; and
instructing, with the computer, an automated warehousing system to adjust inventory levels at one or more stocking locations based, at least in part, on the detected anomaly.