Page 356 - The Handbook for Quality Management a Complete Guide to Operational Excellence

342 C o n t i n u o u s I m p r o v e m e n t I m p r o v e / D e s i g n S t a g e 343

held constant, or an “experimental variable,” which would be changed in
a prescribed way. The results of the experiment were to be analyzed using
all the muscle of a major mainframe statistical analysis package. All of
the members of the team were confident that no stone had been left
unturned.
Shortly after the program began, the customer quality engineering
supervi sor received a call from his quality engineering representative at
the supplier’s foundry. “We can continue with the experiment if you really
want to,” he said, “but I think we’ve identified the problem and it isn’t on
our list of variables.” It seems that the engineer was in the inspection
room inspecting castings for our project and he noticed a loud “clanging
sound” in the next room. The clanging occurred only a few times each
day, but the engineer soon noticed that the cracked castings came shortly
after the clanging began. Finally he investigated and found the clanging
sound was a relief employee pounding the casting with a hammer to
remove the sand core. Sure enough, the cracked castings had all received
the “hammer treatment”!
This example illustrates a category of human error different from the
inad vertent errors described earlier. Technique errors share certain com­
mon fea tures:

• They are unintentional.
• They are usually confined to a single characteristic (e.g., cracks) or
class of characteristics.
• They are often isolated to a few workers who consistently fail.
Solution of technique errors involves the same basic approaches as the
solu tion of inadvertent errors, namely automation, foolproofing, and human
fac tor engineering. In the meantime, unlike inadvertent errors, technique
errors may be caused by a simple lack of understanding that can be cor­
rected by developing better instructions and training.

Willful Errors (Sabotage)
This category of error is unlike either of the two previous categories. Will­
ful errors are often very difficult to detect; however, they do bear certain
trade marks:

• They are not random.
• They don’t “make sense” from an engineering point of view.
• They are difficult to detect.
• Usually only a single worker is involved.
• They begin at once.
• They do not occur when an observer is present.

16_Pyzdek_Ch16_p335-348.indd 343 11/9/12 5:16 PM