Understanding variation

There are four components, the main article: "Understanding Variation" and three linked expansions on the article.

Understanding variation

Activities as varied as answering customers’ questions, preparing invoices, manufacturing products, delivering services, powering offices and factories, and even gaining new business can all be seen as processes. A fundamental property of all but the most trivial of processes is that there will be some undesirable variation in outputs, and the foundation of W. Edwards Deming’s guidance on process improvement is the understanding of that variation, so that it can be reduced.

Control charts are an essential aid. They are widely used in manufacturing, although even there not often in the way Deming (or indeed their creator, Dr. Walter Shewhart) intended. In the service industries they are still little used, which itself explains much of the current difference between manufacturing and service quality. “Improvements” that are based merely on opinion or "gut feel" stand on thin ice. Control charting performance and important related factors – and knowing how to interpret those charts – is the springboard for process improvement. The control chart is the “voice of the process”, telling you both what is actually happening and what the process is capable of doing.

Each time you do your work – be it a service, manufacturing or administrative operation – there will be some variation. Things are never exactly the same. For customers, too, the process of buying a product or subscribing to a service is not always a smooth one. It is a fact of life that processes don’t always work as well, or take the same time, as before. It is rare that we either experience, or do, a “perfect” job. Products that work fine one day can present nasty surprises the next. That is variability, or variation. It is the train (or bus, or taxi, or plane) not arriving when you are late for a meeting, the promised cheque that doesn’t appear, the Helpline that can’t be contacted, or the last self-assembly nut that doesn’t fit.

While variety in products and services can enrich life, variation prevents your customers from enjoying the full benefit. Variation is nasty: it makes life difficult, unpredictable, untrustworthy. It is associated with “bad quality”. Good quality implies reliability, trustworthiness, no nasty surprises. Essentially, a feature of bad quality is too much variation, while a feature of good quality is little variation.

Understanding variation – the launchpad for Deming’s philosophy – is essential for improving any process. It was Dr. Shewhart's breakthrough in understanding variation during the 1920s that formed the foundation of Deming's lifetime's work. Deming repeatedly attributed much of his most important learning to Shewhart.

-----------------------------------

Understanding variation – the launchpad for Deming’s philosophy – is essential for improving any process.

-----------------------------------

For example, Deming wrote: “To Shewhart, quality control meant every activity and every technique that can contribute to better living ... His book (“Economic Control of Quality of Manufactured Product”) emphasizes the need for continual search for better knowledge about materials, how they behave in manufacture, and how the product behaves in use. Economic manufacture requires achievement of statistical control in the process and statistical control of measurements. It requires improvement of the process in every other feasible way.”

Even today, most people's interpretation of the word “quality” is hopelessly narrow and limited compared with Shewhart's understanding in his great book (published over 75 years ago!). And thus so is their use of control charts – if they use them at all.

Shewhart’s discovery

In the 1920s, the Western Electric Company were developing telephone and related equipment, and investing massively to increase their knowledge and ability. Although their early improvement efforts paid handsome dividends, their progress gradually “ran out of steam”. The people were still working as hard, if not harder, than before. The company was still spending much money, time and effort on trying to make things better. But, despite all the work and resources, their quality efforts were achieving less and less. Shewhart was invited to study their problems, and eventually the light dawned. Deming later explained:

“ ... the harder they tried to achieve consistency and uniformity, the worse were the effects. The more they tried to shrink variation, the larger it got. They were naturally also interested in cutting costs. When any kind of error, mistake, or accident occurred, they went to work on it to try to correct it. It was a noble aim. There was only one little trouble – their worthy efforts did not work. Things got worse. ... they were failing to understand the difference between common causes and special causes, and that mixing them up makes things worse. ... Sure we don't like mistakes, complaints from customers, accidents – but if we weigh in at them without understanding, then we make things worse.”

Not just fail to make them better, but actually make them worse. In essence, Shewhart’s breakthrough was to recognize these two very different types of variation – and their very different types of implications as regards improvement efforts, “control”, capability, and so on. What Deming later called common-cause variation is the routine variation to be expected because of what the process is, and the circumstances in which it exists and operates. Special-cause variation is anything noticeable beyond that routine variation.

Surely, very different actions are called for depending on whether something is routine (there all the time) or exceptional (perhaps just one-off). That's it. Not exactly rocket science!

But how can we distinguish between the two types of variation in practice? By using the tool that Shewhart created for the purpose: the control chart. And that is how – and why – the control chart provides guidance for improvement.

-----------------------------------

But how can we distinguish between the two types of variation in practice?
By using the tool that Shewhart created for the purpose: the control chart.

-----------------------------------

However, despite the control chart’s simplicity (and the crucial part it played in the Japanese quality revolution), in the West it is still little understood more than three quarters of a century later, and rarely used outside manufacturing. Instead, everyone involved in key company processes – and particularly those in charge of them – should know how to keep and interpret control charts (see "Unknowns and unknowables") … and do so.

Stable processes and common causes

Let’s revisit an important sentence from the previous section: “ … common-cause variation is the routine variation to be expected because of what the process is, and the circumstances in which it exists and operates”. The range of this common-cause variation can be computed from process data to provide the so-called control limits (see “Calculating control limits”). If the outputs from the process are comfortably contained within these control limits (and show no patterns, trends, etc.) the process is said to be stable (“in statistical control”).

Individual variations in outputs can have any number of contributory causes. For example, weekly sales figures will depend on how many “hot prospects” were available for Sales to contact, whether a particular customer has a particular need this month, even things like the length of traffic jams on the roads to the office in the morning.

It follows that, while we continue to obtain such outputs, it is illogical and impractical to claim that anything specific “caused” any one particular result: for any such result is the kind of result we know can be produced by the whole system of common causes (Deming simply called it the “system”).

So what?

The previous sentence shows the futility of a widespread practice that wastes huge amounts of time and money for businesses. Innumerable organizations try to explain monthly, weekly, daily (and sometimes even hourly!) differences in accounts, sales, production, and performance of all kinds, when in fact the variation is primarily or wholly due to common causes – the “system”. Worse still, the practice unnecessarily increases stress for millions of workers. Individual below-average results are often seen as justification for aggressive management action; above-average results are seen as evidence of the effects of such action (“I told you it was possible”) – and this lucky result probably becomes the next arbitrary target.

However (unless there is some significant change in the system – good or bad) there is, by definition, a roughly 50/50 chance of the next result being above (or below) average. To illustrate, if you toss 25 coins, roughly half the time you’ll get 13 or more heads, and half the time you won’t – either way, praise or blame is hardly justified. As the American consultant Peter Scholtes put it: “Suppose that, when things are going well, we say to people: 'Good job. Keep it up'. Well, the chances are that we are reacting to the higher achievements in a common-cause system. But, in a common-cause system, when things are going well, there is no place to go but down. The only way that can change is for the system to get changed”. Similarly, when things are going badly (again, in a common-cause system), there is no place to go but up.

So, as long as the measurements lie between the control limits, beware of getting distracted by short-term data or (even worse) individual data points.

Unstable processes and special causes

In contrast to stable processes, unstable ones (“out of statistical control”) show changes in behaviour – which is what we mean by unstable. Vibration from a nearby motor may suddenly influence a machine whose mountings have failed, for example. Or half the department may have been fired in an “efficiency drive”. If the effects of such events are enough to drive the process outside the control limits, they become special causes (see “Six processes”) and are thus identifiable.

So changes in behaviour, be they temporary or enduring, do not just happen: they are caused. And such causes must be different from common causes. For common causes do not produce changes in behaviour; they produce continuing similar behaviour. Unlike common causes, it will be profitable to try to determine and remove the cause of individual special-cause variations – because they can be seen. (Of course, that's assuming the change in behaviour is bad. If it's good, you'll still want to find the cause, to see if it's possible to retain it in the system.)

The difference between stable processes (in statistical control) and unstable ones (out of statistical control) is hence an eminently practical one. If we look for the cause of some particular detail in the data, are we likely to find something useful? Or are we setting off on a wild-goose chase, mistakenly believing that we have seen something important (often referred to as “tampering”, a common hazard in mismanaged processes)? Shewhart called the control limits the “economic limits”. He chose them because his experience showed that they minimized the combined cost of making the two kinds of mistakes when interpreting process data: reacting to short-term data when you shouldn’t, or failing to act when you should.

Monitoring is not enough

Watching out for, and removing, special causes is what we call process “monitoring”. It has a valuable “early-warning” role. It stabilizes processes, and stable processes are predictable: with a stable process (again assuming it does not change materially), we can predict process costs and performance, and product (and service) quality and quantity. By comparison, dealing with an unstable process is guesswork. Until you eliminate special causes, you cannot tell what the process will do.

However, process monitoring is just fire-fighting, and this is nowhere near good enough. If process monitoring is all you are using the control chart for, you are missing out on the main purpose for which Shewhart created it: process improvement. Process monitoring merely aims to reach and maintain a state of statistical control. But that’s only the beginning. The next issue is: is the process capable? That means that, when it is in control, it is capable of providing outputs that meet the customers’ stated requirements.

-----------------------------------

If process monitoring is all you are using the control chart for, you are missing out on the main purpose for which Shewhart created it: process improvement.

-----------------------------------

There is a further issue, though: is “capable” sufficient? “Capable” produces satisfied customers, but Deming and others have spoken of the need to “surprise and delight” customers if you want to maintain and grow your business. This means routinely delivering within one day, for example, when the customer used to believe that three days was the best possible. It means comfortable trains consistently arriving on time (with journeys at reasonable prices). It means new houses incorporating solar heating as standard, and designed to conserve water. All this is down to process improvement. And that – to be blunt – starts with top management.

Businesses provide essential products and services and are central to any idea we have of a modern community. However, the sheer size of their social and environmental effects is placing increasing stress on society. Managers usually make decisions for the short term (and feel it is more than their job is worth to do otherwise). These “improvements” normally do have immediate short-term effects, but they also often actually make the whole system unstable. Short-term profit then turns into long-term loss, both for the company and society more broadly. Real and continual process improvement comes from managing the organization as a system, and following (knowingly or unknowingly) Deming’s “System of Profound Knowledge”.

Deming showed how to build a healthy and expanding business by looking to the long term. There may be a way to transform true business “quality” other than that developed by Shewhart and Deming, but we’ve not found it yet. If managers would listen, Deming’s teachings could transform business. If governments would listen, they could transform society.

References:

SHEWHART W. A., Economic Control of Quality of Manufactured Product. van Nostrand (1931); American Society for Quality Control (1980); CEEPress Books, Washington D. C. (1986). The quotation in the text is from Deming's dedication in the 1980 reprint. The book itself, though full of important material, is not for the beginner.

WHEELER, DONALD J. and CHAMBERS, DAVID S., Understanding Statistical Process Control (2nd edition, 1992). The data in Examples E1 and E2 of Panel 3 come from a fascinating, albeit ancient, case history based on the interpretation of hand-drawn control charts discovered in a little-known Japanese company in 1982. An excellent detailed description of this case study is presented in Section 7.2 (pp 154-183) of Understanding Statistical Process Control. Much of the rest of this book is however written at a rather more advanced level than the two books by Dr Wheeler cited below.

Further Reading:

DEMING, W. EDWARDS, Out of the Crisis. Massachusetts Institute of Technology, Center for Advanced Engineering Study (1986); Cambridge University Press (1988); MIT Press, Cambridge, Massachusetts (2000). Dr. Deming's best-known book but (like Shewhart's book) full of valuable material but not easy for the newcomer.

DEMING, W. EDWARDS, The New Economics for Industry, Government, Education. Massachusetts Institute of Technology, Center for Advanced Educational Services (2nd edition, 1994); MIT Press, Cambridge, Massachusetts (2000). Dr. Deming's final book: much shorter and apparently easier language than Out of the Crisis, but with much depth underlying the simpler words.

NEAVE, HENRY R., The Deming Dimension. SPC Press, Knoxville, Tennessee (1990). An introductory yet comprehensive book, much easier to read and ideal as an introduction to either or both of Dr. Deming's books.

WHEELER, DONALD J., Understanding Variation: the Key to Managing Chaos. SPC Press, Knoxville, Tennessee (2nd edition, 1999). The author is, in our judgment, the world master in this area today. This is a superb introductory book, easy to read and quite short, with very clear illustrations of using and interpreting control charts - sometimes with alarming but valuable results!

WHEELER, DONALD J., Making Sense of Data: SPC for the Service Sector. SPC Press, Knoxville, Tennessee (2003). Whereas Understanding Variation is a uniquely brilliant introduction, this is the book you will need if you then want to become your organisation's expert! Excellent and comprehensive, and packed full of practical guidance.

All these books can be obtained from The Transformation Forum - click [here] for the catalog.

By Dr. Henry R. Neave

Shewhart’s discovery

Stable processes and common causes

So what?

Unstable processes and special causes

Monitoring is not enough

References:

Further Reading:

Information on the web: