Level Up Your Quality Control: SPC Tools for Digging into Data

If your only tool is a hammer, every problem looks like a nail.
 
For quality professionals, it’s natural to rely on the tried-and-true format of control charts to deal with the continuous flow of day-to-day quality issues. Control charts are foundational tools for turning raw data into a visual representation that makes it easy for operators and shop-floor supervisors to spot out-of-spec products and variances in processes. 
 
InfinityQS knows that for most quality professionals, the job isn’t just about staying ahead of the next fix. It’s about learning how to prevent those issues from coming up again—and saving the time, money, and pain of constantly dealing with them. The data that feed those control charts (and produce the alerts you are responding to) are just the tip of the analytical iceberg.
 
Let’s take a look at how you can use an array of charts and multi-level filtering in your InfinityQS® ProFicient® SPC software to get the information
you need to manage quality proactively.
 

Go Beyond the Control Chart: Use SPC Tools to Get Multiple Perspectives

You can use InfinityQS capabilities to get multiple perspectives and troubleshoot the root cause of an issue. First, it’s important to realize that you can—and should—be able to quickly and easily view the issue in question in multiple ways to determine how best to address it.
 
Say we are reviewing a View Data chart for a spindle. Measurements of the part are taken for Location A, Location B, and Location C to ensure every measurement is conforming; however, Location C is showing an out-of-spec data point.

Data Chart Pinpointing Out of Spec Issue

To look more closely at that highlighted value, we look at the control chart that shows measurements over time, focusing on Location C. Now we can see that data point in context of how the process is performing over time. The mean looks to be on target, but the Cpk and Ppk values are both low, and the range is steadily increasing.

Control Chart Showing Range Increasing Over TIme

Next, let’s take a look at a distribution histogram for Location C. This view shows a wide spread across the distribution, and a lot of measurements are hitting the upper spec limit. The statistics show that the expected loss is almost 5% if the process continues producing the way it is now. That’s a large opportunity for saving money.

Distribution Histogram Tracking Out of Spec Measurement

Another view can help us confirm that the issue is centralized at Location C and not affecting the other locations. This standardized group control chart shows a comparison of the averages and ranges of all three locations on the spindle. This chart shows the same range increase for Location C, and confirms that this is where we need to focus our efforts.

Standardized Group Chart Tracking Trends

Finally, a box and whisker plot enables us to compare the measurements we’re seeing for the three locations. This plot makes it easy to see that Location C has a broader distribution of data than the other locations, and the associated capability numbers for Location C are much lower than the other measurement points. 

Box and Whisker Plots Enable Comparative Analysis

These charts provide a variety of ways to confirm where the problem lies. But how do you then determine what is actually causing the problem?
 

Using Multi-Level Filtering to Pinpoint a Cause

Using ProFicient, you can filter data six levels deep. This allows you to look closely at specific data regarding your issue, drilling down progressively to narrow down the potential cause.
 
For our example, we start by filtering data by part, crew, and test.

Multi Level Filtering to Drill Down Into Data

The resulting box and whisker plot shows that regardless of crew, Location C had similar issues. From there, we add a filter to determine whether raw material affects the outcome; we don’t see any differentiation there either.
 
Finally, we filter by a machining characteristic: RPM.
Now we see a strong pattern.

Box and Whisker Analysis of RPM

The lower the RPM of the machine, the higher the standard deviation and the lower the Ppk. As the RPMs increase, the variation decreases and capability increases. Now we have identified a direct relationship—and we can take action to address the issue.
 

Modern SPC Solutions Put More Options in Your Toolbox

In day-to-day operations, it’s easy to use the same tool for every problem. But in manufacturing, not every problem is a nail. It’s very important to use a variety of quality charting tools to find the root cause so you can minimize losses and maximize profits.
 
Fortunately, InfinityQS offers a full box of data visualization tools that enable you to look at any issue from multiple directions—quickly and easily. Take a little time to re-familiarize yourself with the variety of charting features you have at your disposal with ProFicient, and you’ll be able to hit all the nails on the head.
Ready to learn more? Download the free white paper A Practical Guide to Selecting the Right Control Chart, and contact your Account Manager for a demo to help you make the most of the tools you have in your ProFicient toolbox. 





 
Rick Sloop
By Rick Sloop
Technical Services Manager
See Full Bio

QUALITY CHECK DIGEST

Never miss a post. Sign up to receive a weekly roundup of the latest Quality Check blogs.

READ MORE POSTS BY THIS AUTHOR

Take the Next Steps