Enabling Data Scientist Use Cases with Discoverability and Metadata

Technical Paper Abstract

Data science uses statistics and algorithms to extract or extrapolate knowledge and insights from noisy, structured, and unstructured data. Data scientists often don’t know what they are looking for until they see patterns or associations. For this reason, all data can be valuable in analyzing and optimizing a process or system. Many CIP-enabled devices possess a rich collection of data that never gets used in a user’s control program. Furthermore, some of that data is buried in the device and not readily exposed. The CIP specifications provide some mechanisms to make that data discoverable, but more could be defined. This session explores options using currently specified techniques as well as some new proposals for making device data more discoverable and understandable thereby enabling its use in data scientist use cases.

Paper and presentation from the 2023 ODVA Industry Conference & 22nd Annual Meeting

Gregory Majcher, Rockwell Automation