Getting a Return from Your Quality System

Companies don’t know what they don’t know. If you really want insights into the operational aspects of a company, you’ve got to collect data and analyze it. That’s when hidden information contained within data will be revealed. Measuring the right things and periodically analyzing the data with the right tools—these are the keys to generating a return on your quality management system investment.
 

What’s the problem?

I spend a lot of time in manufacturing facilities. When touring shop floors, my hosts typically confide in me their manufacturing problems and issues. I naturally ask, “What is causing the problems?” Responses typically vary from “We’re not sure” to “It’s complicated.”
 
My follow-up questions try to encourage my hosts to share opinions regarding the underlying causes of their problems. At this point, usually everyone has a different opinion, and specific culprits for the causes are hotly debated. But somewhere along the way, the combatants become aware that, even with all the discussion, no one really knows the true cause. Instead, conjecture and opinion dominate while logic, fact, and detail are absent from the dialogue.
 
After listening to their list of potential causes, I typically ask, “How could we be certain that the true cause is X or Y?”
 
Light bulbs go on and they usually happen upon the right answer: manufacturing data is required. Without data, we are left with opinion and never-ending arguments. It’s at this point that the creation of a rational data collection plan seems so, well, rational.
 
In the absence of data, people resort to opinion and conjecture. When we don’t really know something for sure, we tend to fill in the gaps with opinion. What we really need, though, is unbiased information. And that’s where a quality management system can help.
 

The data is there; we just need to do something with it

Imagine your data collection plan includes gathering a few data values each hour. After a few days, you’d have a bunch of data that you could plot, in time-order, on a chart. As a result, you would be able to view the drifts and movement of the data, enabling you to visualize process behavior through time. (This is a good description of a control chart.) These specialized, time-series analysis tools enable you to view process performance and resulting changes over time. It’s the manufacturing equivalent of having a finger on the pulse of a machine—an industrial EKG.
 
So, obviously, we keep our finger on the pulse of the machine to ensure that it’s healthy. When we get information, when we feel the pulse skip a beat (go too fast or abnormally slow), we can react and help the patient. We just need an expert to understand the patient, provide the proper diagnosis, and prescribe the right medication.
 

Machine health and operators

If control charts are a machine’s EKG, then operators are its health care service providers. With the information provided by control charts, operators can quickly make decisions about what needs to be fixed, modified, or adjusted—in real time. Operators are the ones who—with data and information—assess  a machine’s health problems, identify corrective actions, and prescribe the right medication to prevent the same problems from happening in the future. The result is instantaneous fixes, modifications, and improvements that help management get a return on a quality system’s investment. At the heart of the diagnosis, treatment, and long-term health of a manufacturing effort is the operator.
  
Getting a return on your quality software investment also involves quality professionals—Six Sigma teams, managers, engineers (and others). These are the people who want to aggregate and analyze data at a higher level, and do so on a regular basis. They search for trends and valuable information across machines, plants, regions, or even the entire enterprise. When these professional analysts look at summarized data, they distill the information down to a point at which they can tell where their organizations have the greatest opportunities for improvement. Defects, costs, overruns, overfills, underfills, too much scrap—these are the kinds of insights that come from the data that companies gather, aggregate and analyze. Generally speaking, the biggest returns on investment are found in summary data, and if your teams don’t stop and analyze lots of data on a regular basis, you are missing out on the biggest opportunities for cost savings and quality improvement.
 

Costs of quality

If you make a product that can’t be sold, then consider it scrap. That’s expensive. Unlike what I’ve heard many times, the presence of scrap does not have to be considered “the cost of doing business.” I’ve been to all kinds of manufacturing facilities, and in almost every case, I see bins stationed on the shop floor that are filled with scrap.
 
I always make a point to look in the bins and ask a bunch of questions about what I see. I’m amazed at how much manufacturing facilities throw away. It’s truly extraordinary. And it’s amazing how much it costs companies in time, energy. and resources. Surprisingly, it’s common that, when I ask about the bins and their scrap contents, there is a collective shrug of shoulders—an acceptance that scrap is an inevitable consequence of manufacturing. It’s just that I don’t believe it. I believe the presence of scrap is an indicator of an absence of information…and an opportunity for improvement.

Imagine how much more profitable and how much productive these companies would be if they eliminated that scrap…if 3%, 5%, or 10% of their production (and of their space) wasn’t taken up by things they throw away.
 
The “costs of quality” are really the costs of “un-quality.” If a company has bins for all the stuff they throw away, they obviously know they have a problem. But, again, when asked, they typically do not know the causes of the scrap.
 
They know they have a problem, but they can’t pinpoint the underlying issues that generate the scrap. What they need is data that can provide them information for identifying the root causes of poor quality. Many organizations I have worked with seem overwhelmed with data, but starved for information—the information they need to help  transform their quality costs.
 

Who pays?

Who pays for all that scrap? Why, the consumers, of course. The costs of poor quality must be added to the price of products sold in the marketplace. Otherwise, organizations may not generate enough profit to continue operations. Consequently, consumers pay for low quality in the form of increased prices, which are needed to cover the operating costs of organizations who find it challenging to control the amount of waste found in their bins.
 
Frequently, when I challenge the need for scrap bins I hear, “You don’t understand. Making this product is an art. It’s very complex, and we should expect scrap.” Any time I hear a response like this, I am convinced that it is the result of a lack of information and the absence of data.
 

Need the info: quality data

It’s a relatively easy fix. Find out where problem areas exist. Look around for the biggest and most overflowing scrap bins, then ask lots of questions about it. Get the experts involved, formulate a plan for collecting data ASAP. What needs to be collected, you ask? Talk to your operators. Collect the thoughts of your quality professionals and engineers. Then, distill everyone’s collective opinions into a proposed data collection plan.
 
Gather the data, review it with operators, quality operations folks, and stakeholders and see what information can be gleaned from it. If necessary, start the process over again. When diligently performed, data collection plans provide information. These activities will help convert art and mystery into facts and science—information that can help transform business performance and generate the return on investment that you need from your quality system.  
 
Douglas C. Fair
By Douglas C. Fair
Chief Operating Officer
See Full Bio

SUBSCRIBE TO OUR BLOG

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