jump to navigation

Managing and Control Charts July 9, 2011

Posted by Tim Rodgers in Management & leadership, Process engineering, Quality.
Tags: , , , ,

In a previous post I mentioned that my recent position managing a quality engineering department has inspired me to generalize quality and process control theory and extend some of the principles to other management issues. See Process Capability In the Office. I’ve been thinking lately about the similarities between the statistical analysis required to know when a process is out-of-control and needs attention, and the judgement to know when the performance of a person or an organization is trending in the wrong direction and needs attention.

First, some more background for those who are not familiar with process control theory.

Control charts are one of the most powerful and widely-used tools in the quality world, particularly useful in detecting process conditions that require outside intervention. When analyzing control charts it’s important to know the difference between common causes of variability and special causes of variability. All processes are subject to inherent variation due to humans or machines, and there is certainly some value in understanding common causes and reducing this source of variation. However quality engineers are especially interested in finding and eliminating special causes, which are causes that are not inherent in the system and lead to unpredictable or unacceptable process performance.

The actions taken when a special cause is suspected are focused on understanding what has changed in the process, and then taking appropriate action to return to a more predictable state. The danger is when a special cause is suspected but the variation is actually due to common causes. This can lead to over-management of the process and “fixing something that isn’t broken,” often making the situation worse and certainly wasting time. Note that a well-constructed control chart helps the viewer identify those statistically significant changes due to special causes.

Control charts require some kind of measurable variable data, or at least attribute data. Nevertheless the underlying principles of common causes and special causes seem to apply regardless of whether individual performance is quantitatively measured. A person in an assigned job will provide a level of output or performance after a training period or learning curve, and that performance can be expected to vary somewhat from day-to-day or project-to-project, but still be fairly predictable to their co-workers and manager. A manager should be attentive to changes in performance or output that may be signs of special causes that should be analyzed and understood.

However, as before the danger is in over-management (more often known as micromanagement) due to incorrectly interpreting small changes in performance as signs of a trend instead of natural variation. Knowing when to step in and when to stay out is a tricky balancing act for a manager who wants to maximize the performance of the team. By understanding the capabilities and expected outputs of the individuals in the team, a manager will be able to detect unusual shifts in performance and only then take the necessary action.



No comments yet — be the first.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: