Building Parts Manufacturer Saves $8.1 Million in Downtime in 6 Months with AI-Powered Predictive Maintenance from Azima DLI

The company’s team achieves this with the help of AI-powered predictive maintenance software and vibration sensors from Azima, in-depth vibration monitoring with Pruftechnik VibGuard, and precision laser alignment with Pruftechnik RotAlign.

The Azima DLI AI-powered diagnostic engine draws on 30 years of maintenance and reliability data – 100 trillion data points across 50 machinery types – giving the building parts manufacturer the power of maintenance recommendations rooted in a wealth of real-world data.

Since beginning their journey with Azima in late 2024, the manufacturer has saved $8.1 million in downtime costs, reducing downtime by over 637 hours.

Unplanned downtime was costing the manufacturer millions of dollars a year. Additionally, equipment acquisition was a source of overspending, since managers would often buy new assets instead of repairing existing ones.

They sought to implement a predictive maintenance strategy so crews could stay ahead of equipment failure and perform maintenance during planned outages.

The Division Reliability Engineer for the manufacturer works in the Roofing and Asphalt division, which focuses on roofing shingles, components, and related products. He is responsible for helping plants implement and sustain predictive maintenance programs.

The engineer manages the reliability programs at 13 roofing and asphalt plants located across the U.S. He utilizes a combination of Fluke hardware and software – the VibGuard condition monitoring system, RotAlign laser shaft alignment tools, and Azima DLI AI sensors and analytics – to slash downtime.

AI-Powered Predictive Maintenance

Azima has been revolutionary for the manufacturer, providing early warning of failure modes, whether it’s bearing deterioration or a pump or motor developing a fault. The early
warning gives maintenance crews ample time to repair or replace components during
scheduled downtime, avoiding any sudden, unexpected failures.

Azima AI-powered vibration analysis software continuously analyzes asset vibration data, checking for signs of emerging faults. The software’s diagnostic engine discovers vibration patterns that signal coming failures and creates a series of alerts, assigning each one a priority level, along with recommended actions for maintenance teams to follow. After that, the report goes to a human analyst who checks and verifies the prescriptive action plan.

The Division Reliability Engineer explains that because the AI processes large volumes of data so quickly, his team gets a full picture of every problem in the plant, making it much easier to plan repairs effectively:

“Azima allows us to prioritize our maintenance down days… beforehand we were using a lot of reactive work. Now we’re able to predict what’s going to fail and provide better insight and better planning for our down days than what we previously had.”

What Makes Azima AI Different? Recommendations You Can Trust

The Division Reliability Engineer cites the quality of Azima DLI maintenance prescriptions as a key differentiator: reliable, actionable, AI-powered recommendations.

“Azima provides very accurate and precise repair recommendations, which allows our maintenance teams to know exactly what to work on. With some of the other companies, you seem to get generalities and are asked to go do inspections before making repairs.”

AI Trained on 30+ Years of Industrial Data and 100 Trillion Data Points

“You’re able to collect some very high-resolution data with their wireless sensors, that, as an analyst, is what makes you able to perform very complex analysis,” says the Division Reliability Engineer.

The Azima DLI Diagnostic Engine has been trained on over 30 years of data, drawing from 100 trillion data points across 50 machinery component types.

Wireless Sensors and Online Monitoring Deliver High-Resolution Data

Azima wireless vibration sensors are key to the building parts manufacturer’s predictive maintenance program, gathering high-quality vibration data across the plant.

“For Azima wireless sensors, they’re a very robust sensor. Highly reliable… versus other vendors where we see a lot of failure rates,” says the Division Reliability Engineer.

His team synthesizes Azima wireless sensor data with the wealth of wired, online condition monitoring data from the Pruftechnik VibGuard.

“VibGuard is kind of a continuous monitoring system. It not only helps us to find mechanical defects, but process issues as well, because it can integrate back into PLCs.”

To demonstrate the power of VibGuard, Division Reliability Engineer shares an example: his team now knows when adjustments have been made to their ‘cut-to-length’ area, a task more challenging with a traditional vibration analysis tool. VibGuard PLC integration empowers the team to pinpoint the source of the problem and ultimately prolong bearing life.

Boosting Asset Lifespan and Cutting Costs with Laser Alignment

The team began using Fluke Reliability products as early as 2008 with Pruftechnik alignment systems. Currently, they rely on the Pruftechnik RotAlign single-laser alignment system for quick, high-precision shaft alignments.

“RotAlign provides such high-quality alignments. That, and the ease of use compared to other systems, just makes them so much more beneficial than other manufacturers,” says the Division Reliability Engineer.

“What makes RotAlign the best alignment tool on the market, is the ease of use of the interface, plus the ability to do single-sweep mode… along with taking backlash out of the equation to provide very accurate alignments.”

He shares that RotAlign probably extends his equipment lifespan by a factor of 10, with pumps that might have failed after a few months achieving a service life of 3 to 4 years.

World-Class Customer Service and Expert Guidance

The Division Reliability Engineer says one of his favorite things about working with Fluke is the customer service, especially when it comes to complex reliability problem solving.

“What I love about working with the Fluke Reliability team is the willingness to help find us solutions that benefit our company,” he says. “There’s never a time we bring up an issue that we’re having that they don’t go, “oh yeah, we’ve been working on something like this, how does this sound to you?””

Fighting Back Against the Global Expertise Shortage

“We’re seeing a reduction of skilled vibration analysts… and the cost to have a skilled vibration analyst internally is high,” the Division Reliability Engineer recognizes, seeing a connection between the challenges his company has faced and broader industry trends. “By utilizing Azima, you’re going to be able to get a faster resolution than you would developing things internally.”

Azima DLI certified vibration analysts are proven to be 5x more efficient than traditional analysts.

Advice to Maintenance Leaders Feeling Hesitant About AI

For companies who don’t feel ready to take the leap into AI, the engineer explains it’s not as challenging as it seems – and the rewards are worth it.

“Most of us don’t realize we already use AI on a daily basis, whether it’s our phones or our Office products,” Division Reliability Engineer says. “When it comes to predictive maintenance, by embracing that, you’re able to prioritize your workload much more efficiently than how you would without it.

Since implementing Azima six months ago, the building parts manufacturer’s team has achieved a breakthrough in defeating downtime.

“We’ve seen about 637 hours of unplanned downtime avoidance, which for us equates to about $8.1 million in cost avoidance,” the engineer explains.

The system also delivers insights into maintenance processes and helps plan more efficient routines.

The Division Reliability Engineer is looking forward to implementing the eMaint condition monitoring platform to centralize data and reporting from Azima and other sources. He anticipates Azima will deliver even more detailed insights in the future as the AI engine continues to improve through more and more years of exposure to plant data.