Consistency: Constant Versus Random (and Averages)
From the Latin, consistent means “standing firm.” Today, the word is synonymous with stable, constant, regular, or lacking in deviation.
What is of interest to the quality engineer about consistency is that it allows things to be measured and even predicted. There are things about the natural world that are consistent and, therefore, can be measured and predicted. The movement of the earth is so consistent that we have been able to measure its movements to define the date and time. We have even reached a point where an app on our phone can tell us the precise time the sun will rise in the morning. Knowledge, observation, and technology allow us to measure the natural cycle, more commonly known as “the weather.” Doppler radar and weather stations have given us a leg up on detecting impending weather events, ones that can wreak true havoc such as hurricanes and tornados, and give us a fighting chance to warn those in the path of the destruction.
However, many would argue that much of nature is too random for the kind of measurement that would allow for prediction. Staying with the weather, I’m sure we have all experienced a morning, privy to a weather report that calls for clear skies all day, only to be caught in a downpour at some time that same day, usually leaving us to wonder if “they” can predict the weather at all.
Another natural phenomena—people, or more accurately, human behavior—has been said to be impossible to predict. Essayist Aldous Huxley once said, “Consistency is contrary to nature, contrary to life. The only completely consistent people are the dead.” But, much like the weather, we continue to try. Take the behavioral science unit of the FBI, which is charged with the study of the criminal psyche and whose work, in part, develops “profiles” that aid in the capture of wanted criminals.
And anyone that has ever played fantasy sports will realize that we also have strived to predict human performance, applying statistics and past performance to help predict which athletes will perform the best in the next game or upcoming season.
Analyzing player performance fits in with the discussion of consistent versus random when we talk about averages. Often mistaken for a measure of consistency—and it is, in a broader sense—averages can be misleading. A hitter can carry a .300 average (or three hits in every ten at bats) over an entire season, unarguably considered an impressive batting average. However, we may not be talking about a consistency of three hits every ten at bats. It is more likely that we would be looking at “cold streaks” mixed with “hot streaks,” meaning a time of no hits in 30 at bats and a subsequent time of 20 hits in 30 at bats. That is certainly not the consistency needed for a quality manufacturing operation.
In manufacturing, our goal is to create a truly consistent process, one that can be measured, controlled and corrected. And talk about consistent! This month’s Quality includes Hill Cox’s 200th column. So check it out, along with everything else we have to offer.
Enjoy and thanks for reading!