SPC for the Real World

SPC, or statistical process control, has been around for a long time. While everyone’s heard about it, not many people are doing much with it.

John Clemons, Director of Manufacturing IT, Maverick Technologies
John Clemons, Director of Manufacturing IT, Maverick Technologies

In short, SPC is defined as the application of statistical methods to control manufacturing processes. The statistical methods part means that there’s a lot of math in the background, but it’s the control of the manufacturing processes that is interesting.

SPC, particularly real-time SPC, can help you avoid failures. By using SPC, you can tell when a process is about to go out of control. And, if you know it’s about to lose control, you can do something about it. This helps significantly reduce quality issues, downgrades and rework. It helps you identify problems before they become problems, and then eliminate— or at least minimize—the failures.

SPC helps you improve product consistency. With SPC you can look at the process at lot closer so you can start ferreting out the causes of product variability. You’ll find that processes you thought were very stable are, in fact, not nearly so. SPC lets you identify those processes, determine the degree to which they’re stable or unstable, and then take actions to improve the process. And SPC gives you the information to figure out if the steps you’re taking to improve the process actually work.

Historically, SPC helps you with your continuous improvement initiatives. Like most people, you probably have some type of continuous improvement program underway, like Six Sigma, lean manufacturing, etc. SPC helps you analyze where you need improvements and helps to pinpoint the specific types of improvements you need to make. And then, as you make those improvements, SPC helps you verify that the improvements are actually working and things are actually getting better—just as you hoped. 

>> Reveal Hidden Assets: SCADA systems can uncover hidden assets and economize others. Visit http://bit.ly/13XcLja

SPC makes sense if your manufacturing processes are generally stable and consistent. If you know your processes are unstable without SPC telling you, then you probably don’t need SPC. Get your processes as stable as you can without SPC, and then apply SPC to take them to the next level.

If your process improvement initiatives have reached a plateau with your existing tools, SPC might be a good way to get them rolling forward again. If you’re already using some tools as part of your quality programs and you’re ready for some more advanced tools, then SPC might be the answer. If you’re already set up with tools like SPC in place, then it might be time for something a little more advanced. SPC gives you a lot of power.

Data is key
The most basic component of SPC is data. You need to have data, and you have to get it from somewhere. Your data can reside almost anywhere, but it most commonly resides in a database, historian or both. How the data got there isn’t really important—whether automatically (which is probably preferred) or manually—as long as you have access to it.

The most fundamental aspect of SPC is real-time data visibility—that is, you can see all data as it is collected. So, even before you start to apply all the SPC rules, real-time data gives you and the operators a lot of important information.

Once have real-time data you can see, you can apply all the SPC rules. You can set up specification limits with real-time alarms and notifications based on those spec limits. Or apply the basic

SPC rules and set up real-time alarms and notifications. You can even set up the pattern rules or run rules with real-time alarms and notifications on them, too. And with all of this displayed in real time, the people who can actually do something about all this—namely, the operators on the shop floor—get to see the information they actually need as it happens. Historical analysis takes the SPC data to a whole new level.

Don’t worry about the SPC math. Instead, concentrate on the business benefits because, in the real world, SPC has some major ones.

>> John Clemons is Director of Manufacturing IT for Maverick Technologies (www.mavtechglobal.com), which delivers systems integration, industrial automation, process control and automation engineering services.

More in Software