The stories of Dr. Deming conducting this onstage were “legendary” – especially when he lit into the executives (in attendance) for their personal performance failure. Proving – that what they have been doing back at work – blaming the workers for systems issues and normal statistical variation – was just as ridiculous.
Proving that Zero Defects is just not possible in a world chock full of variation.
Note: the lip-sync is way off!
From the Videos on YouTube…
The “Red Bead Experiment” was an interactive teaching tool of Dr. Edward Deming. In the experiment, a corporation is formed from “willing workers”, quality control personnel, a data recorder, and a foreman. The corporation’s product is white beads, which are produced by dipping a paddle into a supply of beads. The paddle has 50 holes in it, and each hole will hold one bead. Unfortunately, there are not only white beads in the bead supply, but some defective red beads. The production of the beads is strictly controlled by an approved procedure.
Various techniques are used to ensure a quality (no red bead) product. There are quality control inspectors, feedback to the workers, merit pay for superior performance, performance appraisals, procedure compliance, posters and quality programs. The foreman, quality control, and the workers all put forth their best efforts to produce a quality product. The experiment allows the demonstration of the effectiveness (or ineffectiveness) of the various methods. Some humor is also included along the way.
Describing the Red Bead Experiment has all the dangers of writing a good movie review. One does not want to give out the complete plot line in the description. Suffice it to say that at the end of the experiment, a Statistical Process Control chart is utilized to examine the results of the experiment. What is discovered is that several of the actions taken (which are commonly seen every day in the workplace) were detrimental to the employees and the workplace, and had no improving effect on the process. The concluding comments point out the hazards of misuse of performance data, and how to properly use performance data in a quality environment in order to achieve continual improvement.
Presentation by: Steve Prevette