The Lights Are On, But Nobody Is Home

Traffic light monitoring systems require modifications to be effective

There are many different process control methods and procedures available to the quality practitioner. A popular but problematic visual technique employs the traffic light analogy.

As discussed in the article, “Stoplight Charts With SPC Inside,” by Steven Prevette (Quality Progress, 2004), the traffic light process control system works as follows:
• Specifications for product acceptability are established by the responsible (or in some cases, the irresponsible) authority.
• A dashboard is established with the key process input variables (KPIV).
• Every KPIV has a traffic light.
• If the variable is in the acceptable range, then the light will be green. If the process moves into the warning zone, a yellow light appears, and if the process moves into the unacceptable range, then a red light appears.

…

Want to continue?

By logging in you agree to receive communication from Quality Digest. Privacy Policy.

Create a FREE account

Forgot My Password

Comments

Confused

I guess I just don't get it...why do we need the traffic lights? Are managers really just too dumb to ever hope we can teach them to read a control chart?

Traffic Lights and Management

The gist of my 2004 article was also that traditional setting of traffic light colors based upon arbitrary thresholds will either cause inadvertent reaction to random noise, or missing the detection of a trend when it begins. The article did propose using SPC criteria for the Red and Green colors, where if the indicator was stable, management would need to decide if it was stable at an acceptable level or improvement was needed. See also http://www.youtube.com/watch?v=fXYtlxZe06g, a webinar for the ASQ Stat Division

To wit:

Red = SPC trend detected in the wrong direction

Yellow = SPC shows chart is stable. Management has decided that the stable level is not acceptable

Green = SPC trend detected in the right direction OR SPC shows chart is stable and level is acceptable.

Now, I do agree in principle with Rip that managers ought to be able to understand the SPC chart itself. But, reality intervenes in the following manner:

1. Several managers refuse to understand SPC. Can't do much there, except peer pressure will be brought to bear eventually.

2. Managers aren't the analysts. They don't have time. They rely upon some form of quick hitting, 30 second elevator speech on what is going on with the metrics. At least using Red for a trend in the adverse direction (as monitored on a SPC chart), and Yellow for a stable system which is stable where it should not be, allows managers to quickly go to the problem areas. And the separation of stable (common cause) from trend (special cause) also gives them an indication of what should be done. That is what the analyst is being paid the big bucks for. Do homework for the manager.

3. The Yellow/Green separation allows some corporate "memory" for is the stable system behaving where we want it to - or are we long overdue for an improving trend?

I will say I am a proponent of "progress, no matter how slight, is progress". So, if there is an improving trend, even if not yet in compliance, I prefer the use of Green. Alternatively, an adverse trend, even if the current level is compliant, should be made Red so we may act before it becomes non-compliant.

The current article adds some interesting thoughts to the processing of the data. The question is - where is the right balance between a "quick read" by management (You may want to look at page 31 in the September 2005 QP for input from an actual corporate Vice President on this issue) and giving the manager way too much detail.

Steve Prevette, steven.prevette@srs.gov

ASQ Fellow, CQE