What is Statistical Process Control?

Understand the key concepts and applications of Statistical Process Control (SPC) in manufacturing.

What is Statistical Process Control?

Stockholm, Sweden

1/15/2025

David Leiva

David Leiva

Lean Six Sigma Master Black Belt

In order to clarify the purpose of using SPC to monitor a given process, one could imagine an everyday analogy. Let's say you've decided to bake bread in your kitchen. You gather all the ingredients and get started. You also want to ensure that the results of your baking process (i.e. your bread) conforms to your desired outcomes. But - to do this - what do you measure? Why do you measure it? And what happens if you don't?

Bread

The answers might seem obvious. You might be thinking, "I'm an expert—I know what's too little or too much." But what if it's your first time baking bread? Can you really rely on your instincts if you're not an experienced cook?

The consequences of these differences, whether big or small, can be significant. That's where the work of statisticians and continuous improvement experts like Walter Shewhart and Edwards Deming comes in. Statisticians over the years have worked on a set of principles to help us understand when differences in a process are significant enough to require action. This concept is now known as Statistical Process Control (SPC).

Papers

SPC is a powerful tool that helps you understand how much variation exists in your process, no matter what you're measuring. It shows how consistent (or inconsistent) your process is, which is crucial in continuous processes. After all, consistency is key to reliable results.

You might be thinking, "But I want my process to have absolutely no variation." Is that even possible?

The reality is, even the best processes have some level of variation. SPC helps us determine whether that variation is small enough to be considered normal or if it's large enough to raise concerns. How do we assess this? Through statistics.

One key statistical concept used in SPC is standard deviation. It helps us measure how consistent a process is. In fact, most processes follow a rule: 99.73% of your data will fall within three standard deviations of the mean. This holds true even for natural processes!

SPC incorporates these standard deviations into control limits—upper and lower boundaries that define expected variation. Any data point that falls outside these limits is considered an outlier.

Man writing on paper

In manufacturing, you can track variables like temperature, pressure, speed, and time with great precision. In high-paced environments, measurements might be taken every second. But what happens if something goes wrong?

SPC helps identify outliers before they become problems. The goal is to measure critical characteristics and understand their impact on the process, allowing you to act proactively rather than reactively.

There are various ways to track these measurements depending on the number of samples or the type of data you're working with. The key is to shift from reacting to problems to preventing them. This is why many companies with standards such as ISO, IATF, or VDA rely on these tools, to help them understand how much the process is varying and if they should be concerned about it.

The goal of engineers, statisticians, quality and Six Sigma Professionals is not looking for the perfect processes with zero variability. SPC recognizes there can be imperfect processes, and allows us to see how big this imperfection is. If we are facing new conditions, such as new people, systems, suppliers, etc. we want to know how big is this effect.

Tools like Dominion SPC allow you to monitor changes in real-time. Whether you're producing 100 or 10,000 products per second, these tools can help you determine whether your process is operating within expected limits or if something is outside of those boundaries.

Every process changes, the question then becomes, should you do something?

Imagine you have this change. Should we do something about it? The answer: It depends!

1 day

Imagine that your process experiences the following change from one day to the next

Multiple days

Are we looking at expected process variability?

Outlier

Or, something else?

I'm David, Lean Six Sigma Master Black Belt

David

With over 15 years of experience in various manufacturing industries, I've gained valuable knowledge and insights. I'm happy to help with any questions or comments you might have about this post or statistics in general—feel free to reach out!

david@dominionspc.com

Dominion

©2025 Dominion SPC, Inc.

Socials