Machine health and process optimization applications take AI by the hand
Machine health and process optimization applications take AI by the hand
Machine health and process optimization applications take AI by the hand
Machine health and process optimization applications take AI by the hand
Machine health and process optimization applications take AI by the hand

Machine health and process optimization applications take AI by the hand

Dec. 5, 2023
Augury outlines current applications for AI enhancement and where and how it will scale swiftly.

In the world of industrial technology, we’ve all heard that software is eating the world (famously predicted by Marc Andreessen in 2011). But this takeover by software and associated service-based business models plus changes in how software is developed has laid a perfect foundation for artificial intelligence (AI). So will AI soon be eating the world?

Artificial intelligence has yet to fully transform our lives and work (and it will), but it has made a distinctive and growing mark on manufacturing already, and Artem Kroupenev, vice president of strategy at Augury. 

Process manufacturing and machine health are the two main AI-driven applications, with machine health or predictive maintenance taking the lead, Kroupenev says. “We have found that within the machine health space, especially around rotating equipment, after having built enough of a library of different failure modes on different machines, we can see that there are a lot of similarities that actually benefit from economies of scale and having a very large database,” he adds. To machine learning or artificial intelligence, a pump at one facility is very similar to a pump at a completely different facility or industry.

Whereas machines typically fail in predictable ways, or in ways that are predictable with enough data, process manufacturing has different objectives. Process optimization is more about product quality and is often very specific to each production line. Product seasonality and product objectives greatly influence line processes. “Even identical processes and identical production lines could have different objectives,” Kroupenev says. 

Thus, AI applications need to take those specific applications into account and optimize processes individually. “There are many common denominators across different types of processes, but the way you build the AI solution for process engineering needs to take into account the differences and be flexible enough to be able to provide value across a number of different processes,” Kroupenev says.

Understandably, applications centered around machine health, which are benefitting from a very wide application of data across industries and facilities, are scaling faster than more domain-specific process optimization applications, Kroupenev says. “The goal for us is to create more and more of those commonalities, and then replicate and scale,” he adds.

In general, the adoption of AI technologies is actually moving faster in industries that don’t have a traditional reliability practice or significant automation of reliability and maintenance. With no legacy systems, work can start with a clean slate. “A machine health application actually moves quicker in some cases, for example, in the food and beverage industry versus something like oil and gas, where they have 30 years of reliability practices. There’s a little bit of change management that needs to happen,” Kroupenev says.

For process automation, AI applications are growing but it will take more time. Some industries have already invested work into traditional maintenance methods, and they have the right data and the right measurements around quality and throughput and other important parameters. With some of the infrastructure in place, a process health AI-based solution can work with the data infrastructure that’s already there and build more quickly on that.

Case studies: from route-based to AI-enhanced maintenance

Augury has worked with many companies, small and large, to scale AI applications. The case studies here highlight two manufacturers with wide-scale adoption of Augury’s AI-based services.

For chemical manufacturer DuPont, machine reliability is the foundation of its efficient operations and that wouldn’t be possible without intelligent predictive maintenance. “DuPont is an innovation company—full stop,” says Jim Hunt, corporate digital value delivery leader at DuPont. “The way DuPont is demonstrating its commitment to innovation now is in the digital space.”

Before working with Augury, DuPont’s maintenance and reliability work process was very traditional. It was route based, time intervals, and plans that were frequently interrupted by emergencies. DuPont considered many condition-based monitoring offerings, but what Augury’s AI-based service offered was Level 3 and Level 4 vibration specialists. During the pilot installation, DuPont were getting feedback almost instantly. One of the big, early insights came while baselining one of the first installations, and it detected a motor that was going bad, which the vibe tech couldn’t detect. 

“Knowing that something is going to fail in the future but well enough ahead of time allows you to take orders appropriately,” says Sanjay Rajput, senior reliability engineer, DuPont Nomex. So far, equipment predictions have had 100 percent accuracy.

In another case, for a global paper tissue manufacturer, most of their plants were operating at maximum capacity during the COVID pandemic, with zero room for unplanned downtime. Since working with Augury starting in 2019, artificial intelligence applications have saved more than $23 million in downtime and other costs. That includes nearly 1,000 machine health improvements at an average of 40 improvements per month.

On a weekend in February 2021, one of the plant’s control room alarms for a Yankee gearbox was alerting. Oil pressures and temperatures were increasing beyond the gearbox’s normal range. The team evaluated and decided to wait until the next scheduled downtime. Later that evening, Augury’s AI system picked up rapid anomalies, where all four gearbox sensors were showing a significant temperature increase. Once the maintenance team was messaged by the system, they shut down the machine for investigation, which showed that a corroded shard of metal was blocking the lubrication spray bar inside the gearbox, and the system returned to normal ranges once the blockage was removed. A root-cause analysis determined that the metal shard was associated with an incorrectly installed oil filter assembly during the previous failure.

After the initial success at the test facility, the company decided to roll out AI technology to 30 facilities in 12 countries during the pandemic. Remote training and equipment kits from Augury and the help of local contractors kept the company in operation when it was needed most.

About the Author

Anna Townshend | managing editor

Anna Townshend has been a journalist and editor for almost 20 years. She joined Control Design and Plant Services as managing editor in June 2020. Previously, for more than 10 years, she was the editor of Marina Dock Age and International Dredging Review. In addition to writing and editing thousands of articles in her career, she has been an active speaker on industry panels and presentations, as well as host for the Tool Belt and Control Intelligence podcasts. Email her at [email protected].