“One of the biggest challenges in manufacturing is to get more net good parts off the end of the line,” says Dennis Cocco, president and CEO of Activplant Corp. (London, Ontario; www.activplant.com), a provider of manufacturing software. Cocco admits that there is plenty of information that can be—and is, in many cases—being collected from a variety of sources in order for plant personnel to determine why the appropriate number of good parts isn’t coming off of the end of the line. A couple of the things that are often measured, he says, are overall equipment effectiveness (OEE) and downtime. But he contends, “There’s no correlation between OEE data, downtime data, and where throughput constraints are. We know that for a fact.”
Wait a minute. If a machine goes down, that means there’s no production coming off of that machine: isn’t that a contributing factor to overall production. Not necessarily, Cocco counters. Chances are, there is a buffer so that the machine really doesn’t have an effect, at least not in the short term. He goes on to point out that there are a multitude of interdependencies within a manufacturing environment such that it can be difficult to determine just what really needs to be analyzed. “You can measure tons of data and still not know where your constraints are.” So in order to provide a way to look at an entire plant and determine where the constraints are and how the constrainsts match up so that resources can be directed to the right places, Cocco says they’ve developed what they’re calling a “Throughput Analyzer,” which he describes as a “throughput capability metric that combines the notions behind the Theory of Constraints and the Toyota Production System. It’s a percentage-based metric that quickly identifies, as a percentage, which workcells are the biggest constraints in the plant.”
As is the case with the other elements of Activplant’s manufacturing intelligence software products, the information is accessed directly from shop-floor devices or systems (e.g., PLCs, MES or quality systems). He describes what is necessary for the analyzer as being “a very light amount of data from the automation layer.” Essentially, the analyzer makes a determination of how well an operation is at producing one-piece flow. Constraints are analyzed by four loss categories:
- Downtime (e.g., equipment failure)
- Uptime (taking into account such things as changeovers and tooling trials).
Based on this assessment, the constraints with the biggest impact can be addressed. “This is a very attractive solution for a CFO,” Cocco notes, explaining that by providing an overall look at the constraints within a plant the CFO is able to determine whether investments in capital equipment and/or maintenance are being targeted where there is the best potential returns.
The Throughput Analyzer could also be a great boon to those who would prefer not having to move their production operations to low-wage countries. “You’d be amazed at how many plants are running at the 40% range of what the theoretical maximum output of the plant could be,” Cocco says.
One of the terms he uses a lot when it comes to understanding what’s going on in the manufacturing operation is “clarity.” A word that he doesn’t use as often but could would be “simplicity,” in that the Activplant Performance Management System (APMS) is arranged so that it is not only simple to install but simple to use. A good example is part of what it calls its ActivEssentials offerings, Insight for Microsoft Excel. Cocco explains, “Most plants run on Excel; people like to do their analysis in Excel.” So what they’ve done is create the means by which there is an automated, live link between what’s happening on the plant floor linked to the cells in Excel. This provides the means by which custom reports can be created at as little as half the cost of what’s ordinarily required to produce them. What’s more, Cocco points out that it isn’t necessary to understand the intricacies of databases in order to use the tool. “Our core strength is the ability to take a lot of disparate data from the shop floor and put context to it so that people can understand what it really means,” he says.—GSV