Flooring the competition with real time data

Maintenance supports operations, right? At the best companies, that's the case. Maintenance reports into the operations director or equivalent, and there are agreed, relevant KPIs and supervisory systems to provide data to enable that support. In the very best firms, all that is also linked into the production management systems so that operations and maintenance are aware of each other's schedules and requirements, and so is the business. But for too many manufacturers, that's not the case. For them it's a familiar picture of maintenance versus operations. And that leads to companies, large and small, that consistently fail to live up to their potential – a polite way of saying they waste serious time, money and product in the form of scrap and rework, and are unlikely to be favoured suppliers. In today's globally competitive environment that's not only very bad news, it is unforgivable. Flooring manufacturer Amtico, formerly part of Courtaulds, shows how it can and should be done – and the massive and unexpected additional benefits that can flow from what turns out to be a painless and extremely rewarding exercise. This company, with two near identical sites in Coventry and another in Atlanta, has made the transition from poor to excellent in one fell swoop, on the backs of two individuals and one plant data capture and analysis system. Not only did that system pay for itself in solving its first plant problem, but since then it has opened the door to business understanding and planning that neither of these gentlemen could, in their wildest dreams, have imagined. Amtico – originally American Tile Company, referring to the vinyl flooring process that in the '70s gave Coutaulds its early technical and market edge, but then also the company name following an MBO in 1996 – was a victim of its own success. In 1985, the company was making 1 million yards of flooring per year, and last year that had grown to 4.1 million yards. Amtico was earning well, with the company just churning out product to satisfy demand. So investment in systems to improve production management wasn't forthcoming, much less maintenance management: if it ain't broke, why fix it? As a result, factory management was largely word of mouth; even paper-based works order maintenance systems were only introduced in the last four years. But looking at Coventry's sites, plant had been added to piecemeal over three decades, with all the usual diversity and evolving practice. And each is substantial, with two Calendaring machines for PVC production and three Rotocure laminators, followed by the cutting hall, with a multitude of machines. As Andrew Ross, engineering manager, says: "Because we had grown out of a small company culture, we were still behaving like that. But when you reach a certain size you can't do that and be efficient." He says that having implemented a formal structure of engineering management, attention turned to OEE (overall equipment effectiveness) and the rest, and it became clear that some kind of electronic system was required. "We went to a demonstration of the Mainsaver maintenance management system, and saw MVI's EventsEngine plant data capture and analysis system," he recalls. "Seeing what it could do, we said, 'That's what we need', and we changed our priorities." That decision changed everything. Because by piloting the system on two machines on one production line at Coventry's Factory One, the firm found it could see what was falling between operations and maintenance. Incontrovertible evidence in the form of automatically collected precise plant data, together with simple analytics, brought people and departments together – and built new understanding and thus rapid change for the better. Instant repayment Production engineering manager Jon Cartwright gives the first example. "It was a long-running production issue involving a Banbury, the high shear mixer on a Calendar. For any prolonged shutdown we were getting huge start-up problems costing two or three hours of production, three or four times a week with big material losses. Engineering [maintenance] blamed it on production and production blamed it on engineering." While still commissioning EventsEngine it became clear that Amtico was losing about 10 hours a week and 10s of thousands of pounds worth of product. That focused minds on the issue beyond finger-pointing and led directly to the cause and its resolution. Tests revealed lubrication oil over the product after shutdown. "When the operators shut the machine down and opened the [inspection] door, they should have been turning off the lubrication system, but they thought that would damage the machine." Words were had, and bingo! Says Ross: "Although it sounds simple, you still need detailed information to be able to tie that down quickly, because it's too easy to get stuck in the glue pot. It's only when you have EventsEngine flagging up time after time, you can see, hey, there's definitely something wrong here… Suffice to say, it paid for itself." Then again, Cartwright refers to a problem with a Rotocure, with reports of material dropping off the canvas. "The problem escalated quickly from a production issue to a maintenance issue. The engineers did what they could, couldn't solve it, so it was shut the line down and have a look inside the drums – and they found a siphon tube problem." Without EventsEngine that wouldn't have happened: "We would have limped on – slowed the line down, pointed fingers at engineers." And that, he reckons, would have cost a drop in production of 30%. "The point is, if anybody asks why are you shutting the line down, you need good data." In fact, the system is now used right across production and maintenance management, recording real, detailed events on plant and relating those to operators' and engineers' observations and reports – day and night. Not only are problems identified and resolved quickly but Amtico has found the data invaluable in moving towards best practice across individuals and shifts. "You can't argue with it – it's there it's an absolute record of what the machine was doing," Cartwright points out. But there's more: Amtico's recent upgrade to the system now allows analysis of real production costs right down to each product run and SKU. The system shows instantly which products were affected and to what extent by production problems, the associated costs and gross margins. Says Ross: "We can see product A went through without any problems at all, but product B had a scrap rate of 56%, and that's cost us this… If you're not measuring that data, what you say is 'Last year I ran 4 million yards and I had 10% scrap, so we're not worried.' "Now if you can't do any better than 56%, you'd better have a good pricing mechanism to take that into account. You've got to go from the top to the bottom of how you make your products and sell them. A lot of our products have dropped from our range now because when we do the analysis we say, some of these aren't worth the bother. They're too expensive." All of which raises the question, if this system is so good, why isn't everyone using it? Ross points out that if your plant isn't that big and complex, or your product range is narrow and well understood, or if you already have – and are properly using – good shopfloor data capture (SFDC) systems, you may not need more kit. This plant had no SCADA (supervisory control and data acquisition), no running log for machine speeds and so on. So for Amtico, EventsEngine solved several requirements at once. However, it's worth pondering two points here before you decide either way. First, the automation, precision and description of event recording – and the fact that it can be changed as problems are resolved and others begin to become apparent – makes it popular with operations and maintenance in factories. Their gripes are receiving attention and investment, and that, in itself, ensures take-up and improvement. Second, this is not just about data – it's about easy analytics. Seeing and solving problems buried between functions that no-one has had the time or knowledge to understand, much less resolve, can make a substantial difference – and lead to better decision making that filters all the way up through your business.