“Variation there will always be, between people, in output, in service, in product. What is the variation trying to tell us?
-W. Edwards Deming
Several years ago I had the opportunity to attend an hour-long lecture by Stephen Hawking at Caltech. He returns to Pasadena every summer for a one month retreat, a ritual he started in the 1970s. Several thousand attendees, sitting in both a lecture hall and outdoors on a lawn area, complete with a giant screen, were treated to an evening of reflection by the legendary Cambridge physicist. His focus was “My Brief History,” offering us a glimpse of his life through a twist on his treatise, A Brief History of Time. His introspective presentation revealed his genius, his humility, his search for black holes, and his passion for life, not to mention his dry sense of humour. It ended with questions from three Caltech students, the last of which came from a post-doctoral student, an inquiry Hawking had likely tackled many times before. He relayed the story of an unnamed physicist who once compared himself to both Isaac Newton and Albert Einstein, each placed on a scale of one, lowest, to 10, highest. With this context, Hawking was asked where he would rank himself. While I do not recall the relative rankings posed in the query, I will never forget Hawking’s abrupt reply, “Anyone who compares himself to others is a loser.” -In reference to Dr. Deming, “Variation there will always be.”
Mindful of this natural phenomenon, Hawking’s reply admits the existence of variation, yet disregards the value proposition of a hypothetical ranking of legendary physicists. “We are all different…but we all share the human spirit,” is a common response from Hawking. Could it be he would prefer to accept both the variation and similarities between himself and others and have us move onward in our lives?
Let me transition from Stephen Hawking to “A Brief History of Quality,” and begin this chronicle with three questions, for which the reader’s answers will provide a foundation for thinking about quality. First, what do you call the person who graduates last in his or her class in medical school? The clichéd answer is doctor, the same title as the person who graduates first in their class, as all have met the rigorous academic and residency requirements. What about variation between doctors and, if so, where does it appear? Meanwhile, “goat” is the designation for the officer who graduates last in his or her class at West Point, the US Army’s military academy. In keeping with the Army’s commissioning protocol, all West Point graduates, whether first or goat, begin their military careers as Second Lieutenants. Whereas the use of the term “doctor” paints all of the medical school graduates as the same, without variation, use of the “goat” label implies not all West Point graduates are the same.
On to the second question, one involving numbers. Which two of these three numbers, 5.001, 5.999, and 6.001, is closest to being the same? While these values need not represent anything other than three rational numbers, they could also represent the measured values of hole diameters in an aluminium casting, the “0 to 100 kilometres per hour” acceleration times of a car, or the bacteria levels in a soap solution. In asking this question, ideally with the numbers recorded on a slide, along a line, beginning at zero, the near certain answer is 5.999 and 6.001. The sole exception was the reply from a 12-year old neighbour. He replied 5.001 and 5.999, adding a devilish grin and this clever explanation, “I know which answer you were looking for and I wanted to be different.” Different he was.
For the third question, I offer a well-practiced thought experiment. Imagine a can of a fizzy drink, filled with to the top, but without a closing cover. Now, imagine a small flavour probe in the can, wirelessly connected to a pen in your hand, used to record a flavour profile on a sheet of paper, using flavour as the vertical scale and time on the horizontal scale. At the moment the can is sealed, the probe provides an initial reading of the flavour of the fizzy drink. From this starting point, what is the expected flavour of the drink over time? That is, does the flavour improve over time, as with a fine wine, or does it decrease over time, showing negative signs of ageing? Or, perhaps, remain constant over time? Other options are it decreases, then increases, or increases, then decreases. All of these responses have been submitted from well over one thousand respondents, with a steady decrease over time as the most popular answer, followed by constant over time.
The answers to these three questions reveal assumptions about how we think, which initiated my interest in Deming’s views on management. In a very simple model, we think in terms of patterns, using both black and white and shades of grey. Hour by hour, we routinely use both modes of thought, as context demands, shifting from one to the other and back again. With no known originality, I refer to the former mode as category thinking and the later mode as continuum thinking. Each is extremely useful. Awareness of the contrast between them could prove valuable. Knowing which is more helpful in a given situation might be invaluable. The answer to the first question, “doctor,” shows strong evidence of category thinking, for one absolutely is or absolutely is not a doctor. Black or white. Doctors represent one of many categories of medical professionals, which can be sub-divided into categories, from surgeon to paediatrician to anaesthesiologist. So, too, can nurses and military officers, yet the use of term “goat” to label the bottom ranking officer at West Point is a distinct signal of the relativeness property of continuum thinking that allows us to perceive variation within a given category.
Category thinking is quite useful in allowing us to organise and simplify, much as we use a file cabinet or file folders on a computer or drawers and cabinets in a kitchen. Once we do so, and place doctors in one category or another, we rely on continuum thinking to differentiate those items in a given file. Might this be our thinking when we seek a recommendation for a heart surgeon, knowing heart surgeons have variation in experience, skills, and performance? Else, we ignore the variation within a given category and treat each item; doctor, officer, managing director, or customer; as an interchangeable bit. Such is the simple logic of interchangeable parts, a concept credited to French founders, including Honoré Blanc, in the late 1700s, and given great compliments for advances in world-wide commerce over the past two-hundred years. Might there be evidence of category thinking and interchangeable parts when our health insurance provider (in the US, not the UK’s National Health Service), in response to the potentially higher fees for our preferred heart surgeon, suggests a less expensive, yet potentially less experienced heart surgeon? Far afield from manufacturing, biologists follow the absoluteness of categorisation logic to make assignments with varied life and fossil forms when using genus and species. Innocently, these assigned terms leave us to think, whether counting Cheviot sheep on a sleepless night or the number of goals scored by Wayne Rooney against Liverpool, that each sheep is the same and each goal is the same, rather than a unique occurrence. What can be said for counting customers, suppliers, ideas, and black holes? Are they interchangeable as well?
What can be said of the thinking behind the second question, with 5.999 and 6.001 as closest two numbers in the set of 5.001, 5.999, and 6.001? While each fits the category of being a number, if not a number greater than zero, there was no other implied categorisation. Absent a defined or implied category, such as the absolute requirements for being a doctor, the thinking behind the selection of 5.999 and 6.001 could easily be explained by the relativeness of continuum thinking and the inference that “same” implies proximity. They are a mere 0.002 units apart, far closer than any other pairing in this set of three numbers, hence the confidence with which these two numbers are the dominant choice, absent the mind of a precocious 12-year old. Meanwhile, the margin between the first and last in their medical school class, no matter the overall disparity, is hastily erased when one adopts the absoluteness of category thinking and labels each a doctor. One might wonder if this haste creates waste, a conclusion reached by Edward de Bono in The De Bono Code Book (subtitled Going beyond the limits of language), which proposes, as a provocation exercise, we communicate in codes (numbers, such as 14.11) rather than words. According to de Bono, “Language has been the biggest help to human progress. But, ironically, language has also become the barrier to its own development. We are locked in to words and concepts that are limited and out of date. These force us to see the world in a very old-fashioned way.” An alternative strategy to the use of codes is to be conscious of our two thinking modes, to think about our thinking.
It might be apparent by now the majority of responses to the third question, regarding the flavour profile of a fizzy drink in a sealed can, are also expressions of continuum thinking. As with the relativeness explaining the prevailing answer to the second question, all but three of over 1000 replies (across the US and UK) to the third question have been smooth, continuous flavour profiles. The three outliers revealed an initial flavour level for a short period of time, followed an abrupt, step change to a lower level, followed by a period of steady flavour. When asked for an explanation of this discontinuous profile, each participant made reference to “a point at which the drink goes bad.” Make that instantaneously goes bad (in zero time), as might be the thinking behind an expiration date for fizzy drinks, dairy products, or industrial chemicals.
Although a very small percentage of the flavour profile replies are indicative of the thinking of expiration dates, one need not look far to see them in operation around us-in a grocer’s shops or at work. It is not to say they are bad, or should not be used. Yet let us be mindful of the actions induced by category thinking; on more than one occasion, I have witnessed industrial chemicals in full use right up to the expiration date, and then banned from use and tagged for immediate disposal with the passing of the expiration date. Only seconds before, the chemicals were freely used. While they may rapidly sour, it is unlikely they instantly expire with a big bang, all in keeping with the sentiment of German novelist Thomas Mann’s observation about New Year’s Eve, “Time has no divisions to mark its passage, there is never a thunderstorm or blare of trumpets to announce the beginning of a new month or year. Even when a new century begins it is only we mere mortals who rings bells and fire off pistols.”
The predominant answers to the flavour profile inquiry reveal the majority of us do not think in terms of sudden changes in the flavour of a fizzy drink. Would the replies be any different if the question’s phrasing replaced the “flavour of the fizzy drink” with the “strength of an industrial chemical”? This is food for thought for subsequent research. The reason for phrasing the question in terms of drink flavour is to shift the participant to a framework they likely have never considered, unless, of course, they are employed in the business. From experience, it is easier for someone to answer questions such as the three I have shared, far from the familiarity of one’s daily work, and use their answers as a mirror to reflect on their mode of thinking in each reply. In turn, these thinking modes also reflect on the prevailing explanations of quality, from zero defect quality to six sigma quality, to quality defined by Genichi Taguchi and Deming. Their ideas are the primary focus of this review, as I have found both to have made contributions that offer explanations for the overall success of the Toyota production system (TPS) that cannot be readily explained by the concept of mass production with interchangeable parts that remains the predominant quality construct of the lean community.
The word “quality” has Latin roots, beginning as “qualitas,” coined by Roman philosopher and statesman Marcus Tullius Cicero, who later became an adversary of Marc Antony. Feared by Antony, his power of speech led to his eventual beheading, but long after he introduced his fellow Romans to the vocabulary of qualitas, quantitas, humanitas, and essentia. He is also credited with an extensive list of expressions that translate into English, including difference, infinity, science, and moral. While Plato invented the phrase poiotes for use by his peers, Cicero spoke of qualitas with his peers when focusing on the property of an object, rather than its quantitas or quantity. Two-thousand years later, when writing The New Economics, Deming wrote, “The basic problem anywhere is quality. What is quality? A product or a service possesses quality if it helps somebody and enjoys a good and sustainable market.” As with Cicero, Deming saw quality as a property.
Long after Cicero and well before Deming, quality as a property was the responsibility of a broad network of guilds, associations of artisans who controlled the practice of their craft in a given region, each with their own revered trademark. They were organised as professional societies, not far removed from the concept of today’s trade union. These fraternities guided the development of textile workers, masons, carpenters, and glass workers, from an entry-level apprentice to a master craftsman. They also extended to include wool, silk, and money changers, each with its own high standard for quality.