![]() |
The Top 10 thin gs an ent erpri se lead er ne eds to k now ab out V A RI A TIO N !Category: Corporate, Organizational Issues, Competition (AC48)Originally Submitted on 10/17/97. Knowledge of variation is one of the most important management tools for a process champion or leader. Variation comes in all shapes and sizes. In our human need to deconstruct reality and separate it into parts for "easy managing," we suboptimize our processes and systems by failing to understand the causes of variation. Understanding variation is crucial to managing any enterprise. 1. What is variation? Variation has been studied for centuries and is not at all, new. However, we have been preoccupied with variation the past few decades as we have begun to unlock the keys to optimum enterprise. Variation occurs as a natural part of any system. As we have begun to realize that managing variation is a schema all to its own, we have started focusing more on what causes it and how to manage it. Variation is merely the expected or unexpected differences that occur from one iteration to the next. 2. ALL variation has a causal agent! Variation is caused...by something. Often in discussions involving complex adaptive systems (CAS), we find that cause and effect are often so dislocated as to appear to not to be presen. This transparency complicates many of our issues but the variation is due to some factors, even if they are unknown. 3. Common Cause Variation Common cause variation occurs as a result of many factors, many unexplained. It is the proverbial S___ Happens! cause. It is very difficult sometimes to do anything about this kind of cause. In finance this type of variation is referred to as "market risk." The shift in variation occurs randomly and without warning. This common cause variation is referred to in the quality literature as "system risk." It is often impossible to link common cause variation to any particular event or source. 4. Special Cause Variation This type of variation is responsible for causes that induce variation over and above that inherent in the system. Often, special cause variation occurs as a spike or an exception and is easily identifiable unlike common cause variants, which are inherent shifts in the system and often disguised. 5. TAMPERING! This should remain CAPITALIZED in your mind's eye. This type of variation is caused by "fixing" the wrong problem! These unnecessary adjustments create ripple effects throughout the system and are expressly created by someone "messing about." Often this messing about is created by people trying to react to common cause variation, thinking that by deconstruction and simplification, they can "manipulate" the system to offset common cause. It ain't so and often more damage is created by doing something wrong, then by doing nothing and continuing to learn about system variables. 6. Structural Variation This type of variation occurs as a result of structural organization, or reorganization. Seasonal patterns, industry shifts, consumer buying patterns, regulatory issues which change the landscape, and other cyclical trends and patterns are responsible for structural variation within a system. 7. KNOWING...what is what! Probably the one point that you must take away from this little exercise in variation is that to be able to differentiate between the different types of variation is a key success factor when dealing with processes and systems. If you can identify what is causing the variation, or even what is not causing the variation, you can respond to the system in a way which optimizes the result. 8. Strategies For Dealing with Causes of Variation Strategies for special cause are simple. GET TIMELY DATA! Observe and measure the process in real-time and identify problem signals with accurate metrics. Find out about the issues that are creating bad causes and prevent them and discover what is creating good causes 9. "Controlling Variation!" First of all, it is an oxymoron. It is not meant to say that you can control variation although you can respond to it or prevent it in many cases. What controlling variation is meant to describe is a schema where process owners are empowered and enabled with a set of tools to respond to variation. This "expertise" allows them to become authorities of their own. In other words they can fix problems themselves as experts because they have a system in place which provides them with accurate data with which to make evaluative decisions in order to optimize the process. This is usually done statistically with "control charts." Calculating, studying and responding to these processes with metrics allow process owners to predict future performance of the process and make interventions where needed. It also provides them with feedback on innovation efforts and the capability to respond to problems while the process is operating, truly making them the authority! 10. Recapping Variation We learned that variation has four basic causes. We have discussed various responses to each of those causes and we have made it clear that process owners can become their own experts by learning how to "control" variation, respond to problem signals and predict future performance. What we as leaders need to clearly understand is that we need to find ways to facilitate this process. We must make information availability, communicate responsibility, enable expertise and empower process champions to be just that--champions! The power of leadership must be transferred from taking charge to creating space for leadership to occur. Understanding variation and the necessary means to respond to that variation is a critical first step into that space. G oO D lUck !
This piece was originally submitted by Mike R. Jay, Executive Coach, writer, consultant, who can be reached at variation@leadwise.com, or visited on the web. Mike R. Jay wants you to know: You too can make your organization optimal by understanding enterprise optimization. For more information, drop by our website! The original source is: Variation, Management and W. Edwards Deming, Quality Progress December 1990, pp 29-37, Joiner and Gaudard. |