Common and Special Causes of Variation
Crash Course in Statistics
All businesses or organizations are made up of systems. These systems may be the mixing of ingredients, assembly or any other process that is data driven. When one of these systems fails it’s important to know whose responsible for the fix, and who should keep their mouth shut. There’s actually a science behind determining the blame. It has to do with a nifty tool called control charts which map out individuals points of data.
Control charts are important to process control. Using advanced statistics they can determine whether or not data is acting normal - as in the system has little variation between data points. In the perfect situation you’re processes would never encounter an outlier, but on the occasion that there is an anomaly then you must take the appropriate steps to prevent it from happening again. But whose at fault? Is it the management or the employee that should be held accountable for a failure within the system? This is the juicy bit. It’s all about the data. Decades ago a man by the name of Walter Shewhart first started thinking about this concerning issue.
Walter Shewhart was an American physicist, engineer and statistician, sometimes known as the father of statistical quality control.
In the 1920’s Walter Shewhart developed the idea of the control chart to help decide when the output of a process was part of “a stable system of chance causes”, or whether there was an “assignable cause”. Shewhart described a stable system as one whose variation occurred as a result of small fluctuations (common cause variation).
The observations in a stable process could be viewed as a probability distribution. A stable system that contains no outliers is considered to be in a state of statistical control, or simply, in control. In contrast, an unusually large deviation suggests that the system has been disturbed. Therefore, there was an assignable cause for the disturbance - the system is considered out of control, or unstable.
One Step Further
W. Edwards Deming
Edwards Deming was an American statistician, professor, author, lecturer, and consultant. His contributions to process improvement, statistical analysis, human resources and Six Sigma are numerous. His efforts in manufacturing and business propelled Japan into a technological giant after WW2 and redefined how the car industry operated. His later years were spent teaching and assisting businesses with quality control and statistical measurements.
Deming substituted the term special cause for assignable cause. Deming said that uncovering special causes was the responsibility of the local work force (those who had day-to-day contact with the process). Common causes were part of the system. The system is the responsibility of management. If the common cause variation is too large, it is the responsibility of management to change the system. Deming, stated that 85% of the problems with processes were system problems; later he increased this to over 94%, based on his own experience.
Common causes are dependent on the top level management to solve. These problems are usually too large in scope for the average worker to handle. They require resources, time and decision making to solve. You can identify common causes by examining the outliers in your control charts. Any data that is not statistically normal might be an indicator of a much larger problem such as poor design or working conditions. Managers should be held accountable for special causes. Unfortunately, most managers don’t understand the difference and the worker takes the blame for something he can not control.
Examples of common causes of variation:
- Inappropriate procedures
- Poor design
- Poor maintenance of machines
- Lack of clearly defined standard operating procedures
- Poor working conditions
Special causes according to Deming are the responsibility of the worker. Such problems as the machine breaking down or a malfunction are within their jurisdiction to fix. Workers should be allowed to freely address special causes without having to go up through the chain of command for approval. Natural variation in a system that does not fluctuate outside of the control limits is considered normal. These events will happen with any process and should be dealt accordingly.
Examples of special causes of variation:
- Operator absent
- Poor adjustment of equipment
- Poor maintenance of machines
- Operator falls asleep
- Faulty controllers
Some authors regard this sharp delineation between special causes and common causes, workforce responsibility and management responsibility, as overly simplistic. For example, when a special cause is signaled, and its cause found (rarely an easy task), the local workforce may not have the authority to fix up the problem. Nevertheless, the distinction between special and common causes of variability is a useful one, recognizing that the workforce is accountable for sporadic problems and the management for system problems is an important one.