Wanted: Actionable Information to Make Meaningful Improvements

January 16, 2019
8 min read

I’d like to talk about information. It’s everywhere. The facts you need to make better decisions. The data you need to bring organization to chaos. Sources of information. Uses of information.

But, today I want to talk about the actionableinformation that manufacturers—and, in particular, the quality management folks who work for manufacturers—need to better control and improve their processes, save money, and increase revenue for their companies. To begin, let’s talk about quality control charts.

Ye Olde Quality Control Charts

We all know about quality control charts. They’re the visual tools we use on shop floors to study how a process changes over time. Quality control charts are used to help identify significant process changes that, if not addressed, could result in off-quality product being produced. They are used as preventative measures as well as for problem-solving. We use them to assess process stability, analyze patterns in data, and to reveal previously unknown information.

Quality control charts can show us the actionable information we need to make meaningful improvements in our organizations’ processes and products. Allow me to explain…

A Trip Down Memory Lane

In my early days of quality, I worked for an aerospace manufacturer. The way we gathered data back was different from what you’d expect on today’s high-tech shop floor. The operators took measurements, of course—but mostly to ensure that the parts they were machining were in-spec. Initially, the data weren’t even written down. I’d characterize their actions as more of a “spot check.”

Sometimes operators didn’t even know if a part was rejected until it was inspected by the quality management folks. The quality management staff was gathering data in earnest, though. And they performed all kinds of checks on the machined parts.

A Key to Quality Management: Get Everyone Involved

I suggested that if the operators were going to take a measurement then perhaps they should go ahead and write it down; that way, we could keep track of things. Ultimately, I wanted the operators to review the data and make their own determinations as to what information they could extract from the data they collected.

After about a week or so, we gathered up their data sheets, and I asked them “What are the numbers telling us?” I remember them studying the numbers for a short while, and then one operator said, “Well, it looks like most everything is in-spec.”

Well, needless to say, that was not enough for me. So, I persisted, allowing my idea to come to full boil. “What information do you get from the numbers?” This time I didn’t get much of a response.

So, I finally added, “Let’s do something different. Let’s create a picture from the data.” Indulgent smiles all around. No commitment or concurrence. But smiles. I was anxious to see where this would lead…

A Picture is Worth a Thousand Words

Here’s what I did next. I took one set of data and plotted the values in time sequence on a very simple X-Y graph. We had about 20 data values represented.

“Okay,” I began, addressing the now curious operators and showing them the graph. “These are the same numbers you just gave me. Only now they’ve been plotted on a graph. Now what does the data tell us?”

The smiles turned to grins, and the light bulbs came on. They started to see what I was getting at. They could easily see the data highs and the lows. As a result, they were able to visualize machine variability for the machined feature and they could see how the variability changed over time. Making a picture of the data made all the difference.

From Point A to Point B

We continued chatting. Looking at the graph, I estimated where the average was and, with my pencil, drew a rough horizontal line that approximated the middle of the plot points. I stated that this line represents the average of the 20 data values. Now it was easy for them to see the average and compare its horizontal line to the individual data values. I asked them what they thought this might tell us.

It was quiet for a bit, but one operator spoke. “Based on this chart,” he responded, “I see that the average is quite a bit lower than the feature’s nominal that is called out on the blueprint. That means I’m cutting too much material off of the workpiece. That’s unnecessary. So, I could adjust my CNC program’s offset by a few thousandths of an inch, and the result would mean that the feature would nearly match the nominal value. And if I could do that, then I could take off less material. And that means that my machine would require fewer passes over the workpiece. And if I run fewer passes, then that would minimize wear and tear on the spindle. And that means I wouldn’t have to change cutting tools so often or replace them as quickly.” All of that information from just a simple graph, data values, and an average. We were all very pleased.

Basically, they could make adjustments to a machine and improvements in their work activities based on the actionableinformation obtained from a simple chart. That, for the people standing in that room, was the beginning of something great: The creation of quality control charts to generate operator-actionable information on the shop floor.

What I’m hoping you take from this little story is that the data this group collected started as just numbers on a piece of paper. Just numbers. That’s not information. Numbers are just data. To get information, you need a visual. It’s the picture that provides the actionable information. And, if you don’t have actionable information, then you are not empowering or equipping your operators to continually improve.

So, that’s what I was doing with those operators some 30 years ago. And I think I understated the situation when I said, “A picture is worth a thousand words.” In fact, what I witnessed was that a picture of numbers is a story. It can tell us facts. It can communicate different sub-stories over long periods of time. And these data stories can even stymie us with mysteries that require engineers and operators to work together to solve. But you’ll never get a chance to enjoy the stories (or drive continual improvements) if you don’t make pictures of data.

A Data Set is More Than a Data Set

I’ve never encountered a data set that didn’t have a tremendous amount of information contained within it. Never. I mean that. This leads me to another story. A story with a similar theme.

A short while ago I performed an extensive data review for a current customer. They handed me a spreadsheet and said something like, “We’re interested in what you think about this data.” Pretty vague, maybe even a little mysterious, but I’m always game for this kind of thing.

They gave me three different data sets. I imported their data into our software and proceeded to evaluate, compare and contrast all of the data, identifying what was similar, what was different, and then drawing conclusions based on viewing it all graphically.

The information I uncovered was presented to the company’s 80+ quality managers from around the globe. You could have heard a pin drop as I revealed to them the amazing amount of information contained in their data. They were stunned with the number of insights found in their data—all revealed by making charts and graphs, and by visually representing a relatively small number of individual data values. It was immensely helpful to them, and fun for me.

Results Matter

The shop floor of any manufacturing facility is full of intelligent, capable operators who can work magic with the machines for which they are responsible. They know every nuance and detail about their machines. This enables them to perform very complex manufacturing tasks. They are the Swiss army knives of manufacturing.

Yet, even the most expert of experts can benefit from additional information. This is true with operators. Making pictures of the data they collect enables operators to glean insights into their manufacturing processes that many never knew existed.

Share the Information

In summary, combining the expertise of an operator with graphical representations of data can result in the actionable information you need to make big improvements in quality and your bottom line. It’s not just about the data, and it’s not just about the operators. It’s the combination of the two—along with the actionable information that is garnered from their fusion. Let operators combine their expertise with control charts and the end result is a recipe for great improvement at any organization.

Find out more about obtaining actionable information on our website: