Let’s Talk About Compliance – Part 2: Good Quality Data

For part two of our compliance blog series, we’re discussing the importance of measuring and understanding the compliance of your products and, just as importantly, your processes. As Eric Weisbrod, InfinityQS VP of Product Management, discussed in the introduction to this series, we're going to focus on a general “down to earth” look at compliance: how compliance fits into the day-to-day operations of your manufacturing environment.
 
As a manufacturer, you're only going to produce and sell a product that you can ensure is “good”—and that can mean a lot of things. You measure your products and processes against various standards with a wide array of tools and procedures. As Eric mentioned, what it means for your data to be “good” depends on what you're measuring against: maybe you're checking that “all values are in specification,” that you have “low defect rates,” that a process is showing “no statistical violations,” or that you’re following all local and international regulations. Let’s break that down here, in part two of our series.
 
When I look at compliance, I consider both product and process quality. Finding defective products and stopping them from being shipped is one thing, but it's important to put process at the center of your quality strategy. If your processes perform reliably every step of the way, you can be confident in what you're shipping.

Product Quality

No matter how you look at your data, the end product is always the focus: “does the collected data meet the requirements for this product?” It’s a simple question, right? Yes or no. But there are huge amounts of information that go into answering that question.

Sample Box & Whisker chart: Product quality by feature, production line, and component—standardized to each component’s spec limits
 
It varies by industry, but it all comes down to what I would call “customer-driven” specifications and regulatory concerns. These include things like net content regulations (for filled containers sold in stores, such as beverages) and government standards for safety and label accuracy.

Net contents can be tracked by average and/or minimum weight, and these are heavily regulated. These include the European Union standards of Tolerable Negative Error (TNE), and T1 and T2 set points; and the U.S. Dept. of Commerce’s Maximum Allowable Variation (MAV).
 
As my colleague Jim Frider, InfinityQS Senior Product Marketing Manager, stated in his blog about net content control, “Regardless of implementation, regulations from these entities are designed so that packages are not underfilled, protecting the consumer. Violating these regulations can result in fines or legal actions.”
 

All of the same concepts of net content control apply to specification limits you use in day-to-day operations. Spec limits generally come from customers and/or your engineering department, though helpful feedback can also come from quality personnel and from experienced operators.
 
Suffice it to say that, depending on the industry you’re in, and depending on what you're making, you’ll need to hit a variety of targets in order for you to sell your product. Ensuring that the data you collect is good is therefore paramount to your success.
 

Defects

Not everything in quality is measured on a variable scale between two spec limits. Another aspect of compliance that Eric mentions in the introduction is “low defect rates.” Defects will, and do, happen. It’s a fact of manufacturing life. Keeping those defects to a minimum is what we’re focusing on.
 
Your first step is to find where the defects are coming from, which was the topic of a blog from InfinityQS VP of Statistical Methods, Steve Wise. “Once you find that, then you look at all the inputs into that process that are producing those defects. What's the defect rate? What's the nature of the defects? The variation in the inputs quite likely are what's causing the defects on the output. Often, once you find out where the defects are coming from, it's not so difficult to know how to fix it. But sometimes it is difficult. Sometimes we know exactly where things are coming from, but we have no idea what's causing it. So, you’ve got to dig deeper."
 
"You must look at all the process feature streams that are coming into the process, analyze those, and see which of those aren't set properly or have too much variation, and are creating those defects. And SPC is the tool to do that.”

As we all know, Six Sigma is a set of techniques and tools designed for process improvement. Part of the Six Sigma vision is for manufacturers to have below 3.4 defects per million opportunities (DPMO). That’s the gold standard. In real life, not every process is capable of that level of performance, and not every process requires it, but achieving a consistently low defect count is the goal of attribute-based data analysis.
 
It’s how many of us think about compliance.
 

Lots and Batches

Depending on the type of process you’re running, the checks you make on specifications can vary wildly. You may be checking batches (or lots) or individual parts for variations. Manufacturers are generally mass producing a product (be it cans of soda, electronic components, bags of snacks, or whatever), so you can’t realistically inspect each and every item. So, you perform a version of lot sampling.
 
You look at an entire batch and check for certain features and variations and determine what amount of variation you can tolerate for each batch. Your data might be based on an entire shift, or even an entire day, of measurements. You may have internal or external audits to prove that your products are in compliance. You want to make sure you’re measuring things the right way, and that you’re accountable all along the process. So, what I’m getting at is this: not only does the product have to be in specification, but you have to prove it is; and you might have to prove it to your customers. You might even have to prove it to state or regulatory boards.

Sample Lot Genealogy: Batch traceability as seen in our upcoming Lot Genealogy report

Process Quality

Manufacturers are only as good as their processes. Your organization is judged by the quality and volume of products that your processes produce.
 

In Control

What does it mean when your processes are “in control?” After all, “control” is an integral part of statistical process control (SPC), so it must be important, right? SPC can help you maintain manufacturing momentum, keep you on the path of continuous improvement. So, yeah, it’s important.
 
SPC is used for maintaining process control in the day-in, day-out rigors of making sure that the processes we use in manufacturing are doing what they need to be doing. And they all require constant monitoring. Why do they always need to be monitored? Because they’re machines. As Doug Fair, InfinityQS COO, discusses in his blog, Don’t Drive Blindfolded—The Importance of Real-Time Data in Modern Manufacturing, manufacturing machines have “personalities” all their own.
 
No two machines are alike. Even “identical” machines purchased from the same manufacturer. “We found that under the same test conditions, and for the same material types, each machine produced a little different quality from the others.” So, we constantly monitor them. Because you just never know. "Early warning detection is everything…in automobiles, in healthcare (don’t forget your annual check-up), and in manufacturing. And SPC is all about early warning detection."
 
So, to summarize, to keep things under “control,” we monitor all of our manufacturing machines constantly, looking for any statistical violations.

Statistical Violations

As Steve mentions in another of his informative blogs, “SPC is the Great Correlator. Many of the tools of SPC are exception-based triggers. SPC tools hunt for variation, and when unexpected or excessive variation is found, [your SPC software should then] alert the user. You’re not wasting your time combing through stacks of graphs and charts to look for issues. Let the software tell you. SPC provides evidence of whether something has changed, or something has not changed.” Your decision is then to do something about it or do nothing.

Sample Pareto chart: Breaking down statistical events for key performance indicators (KPIs) at a machining site

Checklists

Many manufacturers make use of checklists. These are often to-do lists for your operators as part of their standard operating procedures (SOP). Checklists ensure your personnel are fulfilling all their responsibilities, and that your equipment is being operated and maintained consistently with each shift, each product changeover, and so on. In our quality management software, you can configure checklists with any checks you want performed, and any phraseology you deem appropriate for the task…and, of course, the answers to the checks.

Sample Enact Checklist: these provide accountability and standardization, help reduce defects, and reduce confusion on the shop floor

Documentation

Having documentation is really important to proving compliance. You need to generate reports to show exactly what you’re doing to remain compliant. And to have the necessary documentation, you must have a system in place that collects all of your important quality data.
 
This system is a combination of many things—shop-floor operators, lab technicians, the equipment you use to collect data (inline scales, monitoring equipment, environmental monitors, programmable logic controllers, and much more), and of course your quality management software. This is where our solutions come into play—ProFicient^™ and Enact^®.
 
Our software enables you to collect data from just about any device imaginable, and then stores it all in a centralized data repository. Your data is then accessible anytime, anywhere. The cutting-edge analytics at the core of our software can then be applied across lines, plants, even the entire enterprise. Data backups occur automatically, reducing the risk of data loss. And audits become much faster and more easily managed.
 

In Closing

No matter how you’re measuring compliance, you need to be able to generate reports to show how compliant you are. What’s happening and when. Who is doing what. Again, a report might be for internal quality control, or for a customer. So, what happens with all that data you’re collecting; how does it help you remain compliant?
 
In the remaining articles of this blog series, we’ll look at the other questions Eric posed in the introduction: a) Are my data collections happening…are the collections taking place AND taking place on time?
 
And b) When something is determined to be out-of-spec, are we recording our solution to the problem? Are we putting the lot on hold, or calling the supervisor...or maintenance?
 
Join us for the rest of the series, too.

• Part 1: Introduction.
• Part 3: Data Collections
• Part 4: Appropriate Actions
• Part 5: Compliance Summary Tile