Scribd
Upload
550 views2 pages

Nomograph & Instructions July 13

This document provides instructions for using a cumulative binomial nomograph to assess test results. It includes two examples: 1) Thirty items were tested, one failed, with a minimum reliability of 80%. The nomograph shows a 99% confidence of rejecting equipment with reliability below 80%. 2) A test of 10 items with maximum 1 failure and minimum reliability of 90% shows less than 30% confidence via the nomograph. The true reliability was 85% giving around 45% confidence.

Uploaded by

lukman Husain
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
550 views2 pages

Nomograph & Instructions July 13

This document provides instructions for using a cumulative binomial nomograph to assess test results. It includes two examples: 1) Thirty items were tested, one failed, with a minimum reliability of 80%. The nomograph shows a 99% confidence of rejecting equipment with reliability below 80%. 2) A test of 10 items with maximum 1 failure and minimum reliability of 90% shows less than 30% confidence via the nomograph. The true reliability was 85% giving around 45% confidence.

Uploaded by

lukman Husain
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
You are on page 1/ 2

Instructions for Assessing test Results Using the Cumulative Binomial

Nomograph

Example 1:
TESTS W/O FAILURE
Thirty items were tested; one item failed; Log (1 - C.L.)
n
Minimum acceptable reliability requirement = 0.80 (80%) Log R

Unknown: Confidence (probability of rejecting bad equipment.


That is; the probability of rejecting equipment that has an actual, but unknown, reliability characteristic that is
less than the minimum acceptable).

Instructions
RELIABILITY (or Ph)

1. Mark left scale at R = the minimum acceptable value or the value to be


demonstrated (0.80).
2. Mark the point on the central grid that corresponds to the test results (n = 30
trials, and failures – 1).

CONFIDENCE LEVEL
3. Draw a straight line between the two pints and extend the line to the right
scale. Read the confidence value on the right scale.

Answer = 99 percent confidence of rejecting bad equipment, or 1 percent chance of


accepting bad equipment.

Example 2:

If we go back to our original class demonstration, we can determine the actual


confidence associated with our original test plan (n = 10, r ≤ 1) and the requirement
(R = .9). Setting these points on the nomograph and drawing a line to the confidence
scale, we read less than 30 percent confidence.

Figure A-6. Nomograph of the Cumulative Binomial Distribution


Adapted from Figure 1, “Graphical Determination of Single-Sample Attribute Plans for Individual Small Lots, “by Shaul P. Lodany,
Journal of Quality Technology, July 1971 (by permission of the American Society for Quality Control Inc).
Instructions for Assessing test Results Using the Cumulative Binomial
Nomograph

Example 1:

Thirty items were tested; one item failed;


Minimum acceptable reliability requirement = 0.80 (80%)

Unknown: Confidence (probability of rejecting bad equipment.


That is; the probability of rejecting equipment that has an actual, but unknown, reliability characteristic that is
less than the minimum acceptable).
Note: What is labeled “Confidence level” in the nomograph
is the probability (given a nominal true reliability) of
Instructions getting more than c failures in n trials.

4. Mark left scale at R = the minimum acceptable value or the value to be


demonstrated (0.80).
5. Mark the point on the central grid that corresponds to the test results (n = 30
trials, and failures – 1).
6. Draw a straight line between the two pints and extend the line to the right scale.
Read the confidence value on the right scale.

Answer = 99 percent confidence of rejecting bad equipment, or 1 percent chance of


accepting bad equipment.
Note: If the true probability of a hit really is 0.8,
Example 2: then the probability of getting 2 or more misses
in 30 trials is 99%.
If we go back to our original class demonstration, we can determine the actual
confidence associated with our original test plan (n = 10, r ≤ 1) and the requirement
(R = .9). Setting these points on the nomograph and drawing a line to the confidence
scale, we read less than 30 percent confidence.

In other words, if the true reliability is just less than 0.9, and our test plan is (n = 10, r ≤ 1), the probability of

passing the test and accepting “bad” equipment is greater than 70 percent.

Note: If the true reliability is 0.90, the probability of getting 2 or more misses in 10 trials is ~26%. So there is
26% chance that we would reject the equipment even if it has a true reliability of 0.90 (met the requirement).
In our actual experimental, the true reliability was 0.85. If we draw a line from .85
through n=10 and r=1, the confidence is slightly higher – approximately 45 percent. In
other words, our chance of passing the test and accepting bad equipment is
approximately 55 percent. (Our experimental results should be close to this.)

NOTE: ACCURACY OF ABOVE ANSWERS ARE LIMITED BY THE RESOLUTION OF


THE NOMOGRAPH. FOR PRECISE ANSWERS, SEE A TABLE OF
CUMULATIVE BINOMIAL VALUES OR COMPUTE THE VALUE.

You might also like