Date post: | 18-Nov-2014 |
Category: |
Documents |
Upload: | manvendra-pratap-singh-bisht |
View: | 222 times |
Download: | 6 times |
Statistical Quality ControlDeming’s Funnel Experiment
Deming’s Funnel Experiment
PreamblePreamble
The funnel experiment is a visual representation The funnel experiment is a visual representation of a process. It shows that a process in control of a process. It shows that a process in control delivers the best results if left alone. The delivers the best results if left alone. The funnel experiment shows the adverse effects of funnel experiment shows the adverse effects of tampering with a process through the four tampering with a process through the four setting methods.These methods are referred setting methods.These methods are referred to as “rules”.to as “rules”.
Deming’s Funnel Experiment
It shows that a process in control delivers the It shows that a process in control delivers the
best results if left alone. The funnel experiment best results if left alone. The funnel experiment
shows the adverse effects of tampering with a shows the adverse effects of tampering with a
process through the four setting rules. The process through the four setting rules. The
experiment was devised by Dr. W. Edwards experiment was devised by Dr. W. Edwards
Deming. It described in his famous book titled Deming. It described in his famous book titled
'Out of the Crisis'.'Out of the Crisis'.
Deming’s Funnel Experiment-Rule 1
Rule 1 - Stable system - the funnel's position does not change Rule 1 - Stable system - the funnel's position does not change
During the first setup, the funnel is aligned above the During the first setup, the funnel is aligned above the
target, and marbles dropped from this location. No target, and marbles dropped from this location. No
action is taken to move the funnel to improve action is taken to move the funnel to improve
performance. The results of rule 1 are a performance. The results of rule 1 are a
disappointment. The marble does not appear to disappointment. The marble does not appear to
behave consistently. The marble rolls off in various behave consistently. The marble rolls off in various
directions for various distances. directions for various distances.
Intuitively, we know that this is probably not the way to
get the best results. However, this strategy will give us
some baseline data. We can compute the results using
one of the other rules with the baseline to measure our
improvement. We could also be lucky enough to hit the
target once in a while.
Deming’s Funnel Experiment-Rule 1
Deming’s Funnel Experiment-Rule 2
Rule 2 - Stable system with larger variance - the Rule 2 - Stable system with larger variance - the funnel's last position is the reference for correction funnel's last position is the reference for correction
During rule 2, we examine the previous resultDuring rule 2, we examine the previous result
and take action to counteract the motion ofand take action to counteract the motion of
the marble. We correct for the error of thethe marble. We correct for the error of the
previous drop. previous drop.
Deming’s Funnel Experiment-Rule 2
Rationale:
This rule attempts to compensate for the Inaccuracy of
the funnel. If the funnel drops the marble Off the target
by a certain amount it is reasonable to suppose that
moving the funnel in the opposite direction by the same
amount will improve the results. This rule requires us
to remember the position of the funnel at the last drop.
Deming’s Funnel Experiment-Rule 2
EXAMPLES 1.A worker may be working to make a unit of uniform weight. If the last item was 2 pounds underweight, increase the setting for the amount of material in the next item by 2 pounds.2.periodic calibrations. One checks a meter’s measurement against a known standard, and adjusts the meter to compensate for the error against the standard. 3.action to change policies and production levels based upon on last month’s budget variances, profit margins, and output.4.setting next year’s goals and targets based upon last year’s levels.
Rule 3 - System "explodes" - the original target is Rule 3 - System "explodes" - the original target is the reference for correction the reference for correction
A possible flaw in rule 2 was that it adjusted the A possible flaw in rule 2 was that it adjusted the
funnel from its last position, rather than funnel from its last position, rather than
relative to the target. relative to the target.
The funnel is set at an equal and opposite The funnel is set at an equal and opposite
direction from the target to compensate for direction from the target to compensate for
the last error.the last error.
Deming’s Funnel Experiment-Rule 3
Rationale:
In this case, we use the target as a basis for our
adjustment, rather than the last position of the funnel
as in rule 2. This is probably our only recourse if we
know only the position of the target and the last drop,
and not the position of the funnel.
Deming’s Funnel Experiment-Rule 3
EXAMPLESEXAMPLESWe see rule 3 at work in systems where two parties react to each other’s We see rule 3 at work in systems where two parties react to each other’s
actions. Their goal is toactions. Their goal is to
maintain parity. maintain parity.
1.1.If one country increases its nuclear arsenal, the rival country increases their If one country increases its nuclear arsenal, the rival country increases their
arsenal to maintain the perceived balance.arsenal to maintain the perceived balance.
2.2.If drug enforcement increases, prices rise due to increased demand, and If drug enforcement increases, prices rise due to increased demand, and
drug runners have incentive to go to further lengths due to increased price.drug runners have incentive to go to further lengths due to increased price.
3.3.A drought occurs one year causing a drop in crop output. Prices rise, causing A drought occurs one year causing a drop in crop output. Prices rise, causing
farmers to plant more crop next year. In the next year, there are surpluses, farmers to plant more crop next year. In the next year, there are surpluses,
causing the price to drop. Farmers plant less next year. The cycle continues…causing the price to drop. Farmers plant less next year. The cycle continues…
Deming’s Funnel Experiment-Rule 3
Deming’s Funnel Experiment-Rule 4
Rule 4 - At drop n, set the funnel right over the n-1 drop.
In an attempt to reduce the variability of the marble drops, we
decide to allow the marble to fall where it wants to. We position
the funnel over the last location of the marble, as that appears to
be the tendency of where the marble tends to stop.
Deming’s Funnel Experiment-Rule 4
Rationale:
The objective of Rule4 is to maintain consistent
results. Even if we miss the target The results should
be consistent, since we always aim for the position of
the last drop. If we are off target we can always take
care of it later.
Deming’s Funnel Experiment-Rule 4
1.A common example of Rule 4 is when we want 1.A common example of Rule 4 is when we want to cut lumber to a uniform length. We use the to cut lumber to a uniform length. We use the piece we just cut in order to measure the piece we just cut in order to measure the location of the next cut.location of the next cut.
2.Adjusting starting time of the next meeting2.Adjusting starting time of the next meeting
based upon actual starting time of the last meetingbased upon actual starting time of the last meeting
3.Benchmarking, in order to find examples to 3.Benchmarking, in order to find examples to follow.follow.
4.A message is passed from one person to4.A message is passed from one person to
the next, who repeats it to another person, and so the next, who repeats it to another person, and so forth.forth.
Deming’s Funnel Experiment-Rule 4
Rules 2,3,and 4 are all examples of process Rules 2,3,and 4 are all examples of process
“TAMPERING/HUNTING”. We take action as a result of the “TAMPERING/HUNTING”. We take action as a result of the
most recent result.most recent result.
Rule 2 leads to a uniform circular pattern, whose size is 40% Rule 2 leads to a uniform circular pattern, whose size is 40%
bigger than the rule 1 circle. This is because the error in distance bigger than the rule 1 circle. This is because the error in distance
from the funnel is independent from one marble to the next.from the funnel is independent from one marble to the next.
In positioning the funnel relative to the previous marble drop, In positioning the funnel relative to the previous marble drop,
we add the error from the first drop to the second drop.we add the error from the first drop to the second drop.
DEMING’S FUNNEL EXPERIMENT-CONCLUSION