Answering your questions from Beyond the Line Pt. 2!
Hi everybody! James Newman, Head of Product & Portfolio Marketing at Augury here. On June 18th we shared how an AI-driven factory and Augury’s latest innovations make next-level productivity possible in our special event, Beyond the Line.
There were a few questions we didn’t have time to get to in the Q&A portion of the event that I’d like to to answer here on The Endpoint community. Yesterday we answered the question, “Is there a vision to optimize and prescribe improvements within the industrial programming domain?."
Today let’s answer this question:
How do you manage to collect all the operational and process data which are beyond the mechanical dimension such as temp. and vibration?
We have a dedicated pipeline methodology, leveraging our tech stack built for Process Health, which has already been used successfully to connect process and operational data from multiple systems to create our process models used for Process Navigator. With Fusion Diagnostics, we extend the use case of those data sets to also be leverageable by the Machine Health platform. There is a more technical deep dive discussion that can be had, but in essence, that model is using standard API and integration protocols to integrate data from external sources, combine and clean those data sources, create necessary relationships, and develop an integrated model of the process and operational data sources within the Production Health ecosystem. The types of data collected vary by customer, but in essence, if there is a means, via API, to collect data, we can collect it, on a time-frame consistent with the refresh rate from the data source. We conduct sampling of the data every minute as part of our model process. For sustainability data specifically, if that data exists in a repository accessible via API, then that data is collectable at the rate of refresh of that system. Typically, we see energy consumption as the low-hanging fruit for collection of sustainability data as it is often integrated and collected within the historian/DCS system and is already collected by default.
With the dynamic recommendations model of the new PH Navigator, we sample data via the pipeline once a minute. However, the frequency of creating a new recommendation is based on discussions with the customer. Feasibly, recommendations can come as often as every 15 minutes, however, with most customers to date, they prefer that operators not be notified that often, so typical refresh of recommendations is occurring at around 1 hour intervals. This means that for refreshed data, we tailor the “need” for new data changes to be consistent with the refresh of recommendations.
Stay tuned over the next couple days as we answer the rest of your questions from Beyond the Line! Catch the full event on-demand here.
Comments
-
Thanks for the answering this James!
The value of a Machine Health plaform lies within it's root capabilities for Condition Based Monitoring. The same can be said for a Process Health platform in that it's value lies within Event Based Detection. The power of Process Health comes from its ability to discern multi-variable interractions (3rd, 4th order) and variable dependencies, many cases disguised as the "black box" of process managment. Coupling these two outcomes provides the end user significant insights that might otherwise go undetected until a failure occurs, be it a machine failure, an out of control process or a quality upset event. Imagine the power enabling the operators and workforce to confidently anticipate and the effectivess available to respond and mitigate issues that might otherwise result in another "firefighting" episode.
0
