Robots at work are a good thing. Robots at rest when they are supposed to be working are a bad thing. The latter is what General Motors Corp. is getting the hang of preventing, according to the automaker’s director of global automation.
Speaking at an April automation conference in Chicago, Mark Franks told the audience that GM had avoided 100 potential failures of robots by analyzing data they send to an external server in the cloud, Bloomberg reported. “If we can avoid a disruption in our manufacturing, we can save ourselves a significant amount of money,” Franks said.
It’s a comment every manufacturer can agree with. Assets that are unavailable when they are needed, or equipment that performs poorly when it is available has lost dollars attached to it. For example, even as Smart Sand Inc., a producer of raw frac sand, reported improved second-quarter net income this year, the company acknowledged in its earnings report that the numbers could have been better were it not for, in part, unplanned downtime at its Oakdale, Wis., facility. And earlier this year, ARC Advisory Group reported its belief that the impact of unplanned downtime on revenue and profitability in the process industries has been “vastly underestimated.” “ARC estimates that downtime costs the process industries somewhere in the area of a trillion dollars a year, “ wrote ARC’s Larry O’Brien in a blog post.
Sometimes unplanned downtime is unavoidable. For instance, when hurricanes like Harvey, Irma and Maria sweep through an area, manufacturers have limited actions they can take to keep operations up and going. But in many instances, manufacturers can and likely should do more to keep their assets operating in optimal condition. Several manufacturers and reliability experts share their ideas.
More Than Maintenance
If your company or facility points to the maintenance function as the keeper of all reliability-related duties, it is making a mistake. Research shows that just 17% of unscheduled downtime can be attributed to the maintenance function, says Life Cycle Engineering’s Brad Cunic, senior vice president of its reliability consulting group.
Given that data point, manufacturers should take note of where downtime opportunities occur. “Defects or downtime are interjected throughout the equipment lifecycle,” from the initial equipment design to procurement functions focused on lowest cost versus total cost of ownership, to equipment operators who may be mishandling the assets they run.
“[Reliability is] really a shared function that must be driven from the top of the organization, and that starts with the alignment of key executives,” he says. “Most successful companies we see have reliability built into the culture, not just as a maintenance activity. That’s where you see sustained performance.”
Such is the case at Cargill Inc., a privately held manufacturer whose breadth of products includes animal nutrition, food ingredients, salt and beef products. The manufacturer embraces the philosophy of reliability as a shared responsibility, says Rick Baldridge, corporate reliability excellence process leader as well as a Society of Maintenance and Reliability Professionals board member.
“We strongly believe that production owns the assets, while maintenance owns the capability of the assets. That takes a lot of collaboration,” including from such domains as safety, customer quality and process safety risk management, Baldridge says.
“Really, the word is holistic. The best analogy is ... your own personal vehicle. If you see a need to take your vehicle in for anything that it may need, you are the one that ultimately makes the decision of all of the things that they find. And your expectation, I’m sure, is that they are highly qualified and really understand the health of your vehicle, and are able to translate that into a language that you understand, and help you make the right decision as to what you are going to do,” he continues. “The one thing they will never be ahead of you on is what you expect out of that vehicle, what is your operating context of that vehicle. And that’s why production needs to be the one that owns the assets. There’s absolutely no difference with physical assets in our facilities.”
Such collaboration isn’t easy, which is why Baldridge also stresses the importance of communication. “We have a very comprehensive integrated approach to reliability, and not any single thing is going to get us to our overall objectives. However, if there was a shiny bullet... it would be change communication.”
Create a Solid Foundation
“The more you can minimize downtime, the more you can improve throughput and quality,” said the operations manager of Intertape Polymer Group’s Danville, Va., facility during an IndustryWeek visit to the 2016 IW Best Plants Award winner.
It’s a sentiment echoed by Ron Englehart, continuous improvement practitioner for Intertape’s operational excellence program, the Intertape Performance System. The implementation of IPS revolves around model machines, those identified as the most critical to the business, and begins with machine cleans to improve performance and reliability.
“Foundationally we are looking to engage our workforce, and we do that through the model machine cleans. The model machine clean is a means of getting cross-functional teams together and improving our associates’ training and knowledge about how that equipment operates day in and day out,” Englehart says.
“As most people know, when you are dealing with inconsistency and instability in your equipment, it’s just the tip of the iceberg of the true cost to the operation, so you have a lot of hidden things beneath the surface, whether it be safety, quality, delivery, or morale as a result of inconsistent uptime,” he continues.
Of course, Intertape augments the people side of reliability with technology. For example, at its greenfield sites and for other critical applications and support equipment, the company has implemented SCADA systems and plant apps that look at downtime tracking. These allow the manufacturer to set trigger events and generate reports to indicate where downtime events occur or where there could be reoccurring events. Intertape also employs artificial intelligence software both as a predictive analytics tool and to fine-tune process parameters.
Such technologies at the company’s facility in Blythewood, S.C., have helped boost uptime there from 92% to 98%.
“I need stable equipment in order to produce consistent product on time. If you are not monitoring your equipment for predictive failure, then you are going to be doing it on an unplanned basis, so what we are trying to be able to do is give ourselves planned outages to control uptime,” Englehart says.
“In the end the technology that we are putting in place allows us to plan the downtime on our terms instead of the equipment’s terms.”
Testing New Technology
Using technology to aid asset uptime is nothing new. Computerized maintenance management systems, vibration analysis, ultrasonic monitoring and other predictive maintenance technologies have had a place at the reliability table for some time. What has changed is the amount of data that can be collected (it has skyrocketed), the price of sensors (a downward trajectory), the introduction of artificial intelligence and machine learning — the sheer promise presented by big data, the IIoT and other transformative technologies.
Cargill has dipped its toes in the IIoT water with its condition-based maintenance. Land O’Lakes Inc. is piloting enterprise manufacturing intelligence in its feed business. General Motors has more than tried on technology. According to Bloomberg, about a quarter of the automaker’s 30,000 factory robots are connected to the cloud. That connectivity enables GM to reduce its spare-parts inventory, thus tying up fewer dollars, by detecting when parts are going to wear out, according to the automation chief.
Such advances are likely just the beginning. GE sees the future and it is digital. “When we think about the industrial company of the future, it is still going to build machines, but these machines are going to have the capacity to respond with greater speed and insights than anyone ever imagined,” GE CEO John Flannery said during the company’s recent Minds + Machines event.
And Andy Kopp, technical director at Myrtle Consulting Group, outlines what he says will be the exciting story in asset uptime — the transition from predictive maintenance to prescriptive. What that means is not only will digital tools gather and analyze data, but they will also suggest the best course of action to take based on the analysis and automatically set that action in motion. Heady stuff.
It could be easy to get caught up in all the excitement of new technologies to boost your equipment reliability efforts. Life Cycle Engineering suggests a measured approach.
“Our philosophy is start with the end in mind — the business results that you need and work backward. First, determine what assets have the biggest impact to your value stream and then what failures are most likely to occur in those assets and what is the most effective mitigation method used to eliminate, detect, or remediate those failures. Then, what are the early signs of failures in those assets?” Cunic said. “What critical control points should be monitored and what are the set points or the alarm, and who should receive those alarms? And how do you ensure that the right people, materials, and procedures are in place to respond? Then you determine what data needs to be collected and analyzed and by whom to promote this continuous improvement of the entire process.”