| by Roderick A. Munro, Ph.D.
Over the years, most organizations have used slogans to encourage people, whether they're managers or shop floor workers, to improve the quality of their work: "Quality is whatever the customer says it is," "Quality is everyone's responsibility" and "You can't inspect quality into a product or service" are three common examples. What about your organization? Any company that's been around for a while has collected enough slogans and quality data for a book, or at least a technical paper. You know, actually writing that book might be a good exercise to help CEOs understand the evolution of their companies' culture and quality systems while they've been at the helm.
That evolution isn't al--ways good. Think about it this way: At any point in your organization's history, have the products or services it provides been truly outstanding? I'm not just talking about meeting customer expectations, but been truly exciting. For many organizations, an objective answer would be, "Yes… but… um… maybe not at the moment."
How did that happen? What factors lead to customers extolling your product or service, and what's caused your customers to think of your products or services differently today? Has anything changed in your organization?
This last question is how I start any problem-solving activity. Usually, when a group gets together, it's to create something that works for a particular purpose or need at hand. That's also how many companies start. But what happens over time? How do we start out with a great product or service, a vision of quality that permeates the organization, but over time end up producing lackluster products?
Do you remember the old telephone game? You sat in a semicircle with a group, and the person at one end whispered some phrase to the next person, who whispered the same phrase to the next person, and so on until the last person said the phrase out loud. What happened? Invariably, what the person recited out loud bore no resemblance to the original phrase. Why not? (By the way, if you try this with a group of highly trained and educated engineers during an FMEA training session, they still won't keep the statement straight.) Like that telephone game, the vision and technical approaches to quality have changed over time as they've been passed from the originators, to CEOs, to consultants, to publishers, to public relations people and so on down the line--like evolution, only backwards.
Maybe we need to revisit our quality roots and understand where today's quality ideas and programs come from.
Let's go back to Walter A. Shewhart, the "father of quality." He developed the concept of control charts and demonstrated that common cause and special cause variation exist in every system, and that you could judge how stable a system was with some simple charts or graphs. Ask many managers today if their company's system is stable, and they won't have a clue what you're talking about, never mind being able to use data to demonstrate that stability.
I like to ask people who do understand control charts why they use a sample size of five for typical X-bar and R charts. (Shewhart preferred a sample size of four.) Most people don't know why, but the answer is that during World War II, the U.S. Department of Defense had to come up with a simple way for untrained personnel to measure the quality of the products that they were making. If you take any group of five numbers and add them up, double the sum and then move the decimal point one place to the left, you'll have the average.
By the way, why do we typically check parts once an hour? Because that procedure was simpler for untrained personnel than calculating a true sampling plan, which could have distracted them from actually making the parts.
Today the question should be: "If these two stories are true, then why are most statistical process control software programs still giving us the range instead of calculating the standard deviation?" For that matter, why can't these programs prompt the operator to take a reading, and if that's not possible, perform a measurement automatically, based on the operation's performance history? I thought computers were supposed to help us.
How about flowcharts? A good flowchart should show the basics of what gets done in a system. However, if you ask the top five or 10 people in your organization to independently construct a flowchart of how the organization actually works, what's the probability that the charts will actually match? What if you extended this exercise to include everyone in the organization? My favorite question relevant to this exercise is, "If you could start from scratch using the people you have now, would your system look like it does today?" If the answer is no (which it usually is), then why aren't you, as the manager, moving your organization in that desired direction?
How should a failure mode and effects analysis (FMEA) be used? Any idea of when this technique was first developed? Many people will tell you that the National Aeronautics and Space Administration developed the concept during the 1960s to help the United States get to the moon and back, and it's true that NASA did use what it called the failure modes and effects analysis. But the concept actually was developed after World War II as a means of identifying the risks to a required activity per military standard MIL-P-1629.
What's the purpose of ISO 9001:2000? Many people trace its origins back to the United Kingdom's BS 5750 standard. However, this one goes back to military standard MIL-Q-9858, which Joseph M. Juran discussed in the first edition of his Quality Control Handbook (McGraw Hill, 1951). ISO 9001 was first published in 1988, when quality was an end-line inspection process and everything had to be checked against customer standards prior to shipment. As ISO 9001 evolved, it continued to serve as a minimum requirement of what an organization should do to be considered a contender in the world marketplace.
ISO 9001:2000 represented the rise of a more process-oriented approach, and the standard now requires an organization to think about what really satisfies its customers.
However, ISO 9001 represents the bare minimum for a "quality" company. If you truly want to be the best in your particular field, you must apply for the Malcolm Baldrige National Quality Award (which also started in 1988 after its enactment by Congress in 1987) or one of the state-level Baldrige Awards.
There have been many quality lessons learned over the years that are valuable to revisit. Here are a few:
• We never learn. A couple of years ago one top statistician I know joked about reviewing résumés for a position his organization wanted to fill. He ran across a person who was a quality director for a company he'd worked for 20 years before. The candidate claimed to have worked on, and solved, the same problems the statistician solved when he worked for the company.
• Old techniques do work. Many people of my generation know about quality from the 1979 NBC documentary, "If Japan Can, Why Can't We?" that reintroduced W. Edwards Deming and control charts to U.S. viewers. Today, control charts are rarely used. However, during the early decades of the last century, many organizations proved that these tools work very well.
• New isn't always better. When Japan started on its road to continual improvement, it had only used equipment to work with. Many organizations today make great strides in quality while using old technology, but when they switch to new technology, they get into trouble. Deming's famous stunt was to go into an organization and, once in the plant, ask that all the automatic controllers be disconnected and any person who wasn't needed to feed materials step away from the machinery. That way, he could observe what the machines were actually capable of doing before reinstalling only those controllers or automatic devices that really helped to decrease product variation.
• Take care of your tools. When I was starting out in quality, an old-time consultant told me that, to be good at our job, it was important to pay attention to two things. They're well known--even listed in ISO/TS 16949:2002--but rarely well executed: Cleanliness and preventive maintenance. Most of the time, simplicity will lead to dramatic results. I was told that if companies focused on these two items, more than 50 percent of their problems would disappear. Now that I'm older and have observed many organizations around the world, I can say that it's true.
• Employees have to participate in change. Nearly every manager I talk with immediately states that people dislike change. Do they really? Or do they simply dislike forced change? I point out that if we went to the manager's house and I said that I was going to get it repainted, choosing the colors and methods myself, it might not go over so well. However, if the manager was given a budget and allowed to choose, that would probably work out fine.
• Keep employees informed. Why do so many people like to either participate in or watch sporting events? I contend it's because at just about any point during the event, everyone can see the score. What about when they're back at work? When do employees find out how well they're doing? Is it every 12 months or so at their annual review?
During the late 1920s and early 1930s, Harvard Business School conducted a major study of people who were employed at a place called the Hawthorne Works in Cicero, Illinois. The company was one of the old AT&T plants and at the time operated in conjunction with Western Electric Co.
Many managers have taken classes where they hear about a study of how lighting in a room affects employee productivity, although few connect it with the Harvard study. Remember, Walter Shewhart worked for Bell Labs (another division of AT&T) and had created control charts during the late 1920s, so they were being used at the Hawthorne Works when the study was conducted. Also during the 1930s, a young engineer named Joseph M. Juran worked there, and for one summer an intern named W. Edwards Deming showed up as well.
The study ended up showing less about how environment affects productivity and more about how people interacted with management.
The lesson wasn't lost on the Japanese manufacturers who, after World War II, pored over the literature from this multiyear study. Almost no one in the United States today remembers any of it. The only reference I'm aware of about it that's still in print is Richard Gillespie's book, Manufacturing Knowledge: A History of the Hawthorne Experiments (Cambridge University Press, 1993). Yet, this study is partly what led to the Japanese development of quality circles.
I truly believe that quality should be simple. That's one reason I enjoy talking with people about what they've learned over time, especially regarding quality. Pick any quality tool or system in circulation today. Find out how it was started and what it was meant to do. Then look at how it's used today. You might find, as have many organizations, that it's become much more complicated over time, and that people have forgotten its original intent. Don't fall into the trap described by philosopher and social critic George Santayana: "Those who forget the past are doomed to repeat it."
Roderick A. Munro, Ph.D., is a business improvement coach with RAM Q Universe Inc., and speaks and writes on various quality and productivity topics. One of his books is Six Sigma for the Shop Floor: A Pocket Guide (ASQ Quality Press, 2001), which uses many of the concepts discussed in this article to help explain why and when Six Sigma tools should be used.
|
