Six Sigma–Specification vs. Control Limits

In the eighth chapter of their book Six Sigma:  The Breakthrough Strategy Revolutionizing the World’s Top Corporations, the authors Mikel Harry, Ph.D., and Richard Schroeder discuss “Measuring Performance on the Sigma Scale.”   In the previous chapter, the authors discussed the Breakthrough Strategy of implementing Six Sigma on the business, operational, and process level.

In this chapter they focus on the question “how does improving the Sigma level of a company’s processes improve that company’s performance?”   One of the ways this is done is by setting a control limit in order to control the variation within the speification limit.

One way to explain these two concepts is by using an analogy.   Let’s say you’re in a car that is traveling down the road and you don’t want to hit the leave the road because the shoulder has rocks or, even worse, a sharp drop off a cliff.   One of the ways you can do this is by focusing on staying within the guardrail.   Now, if you’re doing a process that is churning out units on an assembly line, you don’t want to have defective units which can occur if the units are out of specification.   So you measure the variation from the center and you call the point where the variation goes out of specification as the specification limit.   It’s the equivalent of the guardrail in the analogy.   You stay within the specification limit, no defects.   You stray outside of the specification limit, then you’ve got a defect.

If you’re traveling down the road, rather than trying to avoid the guardrail, and even safer method of driving is to make sure you that you stay in your lane (except when passing a car, for example).   The lane line on the right-hand side of your car is far enough away from the guardrail that, if you focus your effort on staying within your lane, you will almost assuredly never be in danger of hitting the guardrail.   Now, if you’re doing the process mentioned in the paragraph above, and you want to stay within the specification limits, then you setup control limit.    It’s the equivalent of the lane in the analogy.   You stay within the control limit, then you end up staying with the specification limit, and there are no defects.

This brings up another reason for the 1.5-sigma shift which is a phenomenon where the long-term performance of a process is 1.5 sigma less than the short-term performance.   The analogy for the car your car’s steering capability.   If you point the car in a certain direction, and then leave off the steering wheel, are your wheels and chassis aligned in such a way that the car will still go in that direction?   Or will the car drift to the left or right?    Now if you are driving along and making sure the wheel is in the same direction, the car will go straight.   That is the short-term steering capability of the car.   However, if your car’s wheel alignment is such that the car will tend to steer to the left or right, it may require repeated inputs from you to keep the car aligned correctly.   In a similar way, the short-term capability of a process may be at 4 sigma, but if measure it in the long run, there may be errors that are causing the quality equivalent of faulty wheel alignment in that they cause the long-term capability of a process to be at 2.5 sigma instead.

This is why it is vital to account for the “shift and drift” phenomena mentioned above by dividing the total process variation into short-term and long-term components.   This is the only way to make sure the process is maintaining high quality, or in our driving analogy, to make sure the car stays on the road!


Leave a Reply

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

You are commenting using your 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 )

Google+ photo

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

Connecting to %s

%d bloggers like this: