US 12,327,118 B2
Industrial internet of things AIOPs workflows
Francisco Maturana, Lyndhurst, OH (US); and Jay W. Schiele, Union Grove, WI (US)
Assigned to Rockwell Automation Technologies, Inc., Mayfield Heights, OH (US)
Filed by Rockwell Automation Technologies, Inc., Mayfield Heights, OH (US)
Filed on Apr. 14, 2021, as Appl. No. 17/230,115.
Prior Publication US 2022/0334838 A1, Oct. 20, 2022
Int. Cl. G06N 20/00 (2019.01); G06F 9/30 (2018.01); G06F 18/214 (2023.01)
CPC G06F 9/3017 (2013.01) [G06F 9/30123 (2013.01); G06F 18/2148 (2023.01); G06N 20/00 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A method of streamlining industrial internet of things (IIOT) solutions, the method comprising:
receiving, by a context analysis component, first runtime data from one or more data sources associated with an industrial automation process;
analyzing, by the context analysis component, a context of the first runtime data, wherein the analyzing comprises defining the one or more data sources associated with the first runtime data and identifying data gathering processes for the one or more data sources;
based on the analyzing the context of the first runtime data, developing, by the context analysis component, an application comprising a collection function that uses the identified data gathering processes to collect data from the one or more data sources and a context function that contextualizes the data by applying a context to the data;
deploying, by a deployment component, the application;
processing second runtime data with the application to generate contextualized data;
based on the contextualized data, developing, by a machine learning operations component, at least one machine learning model;
generating, by a management component, a second application implementing the at least one machine learning model in the application;
deploying, by the management component, the second application with the at least one machine learning model;
processing third runtime data with the second application to generate additional contextualized data, wherein the processing comprises generating, by the at least one machine learning model, at least one output, wherein the at least one output is generated based on the additional contextualized data;
based on the third runtime data and the at least one output of the at least one machine learning model, generating, by a detection component, a first alert and a second alert about the at least one machine learning model; and
transmitting, by the detection component, the first alert to the context analysis component and the second alert to the machine learning operations component.