Browse Prior Art Database

Realistic Random Failure Criterion Certification

IP.com Disclosure Number: IPCOM000040400D
Original Publication Date: 1987-Nov-01
Included in the Prior Art Database: 2005-Feb-02
Document File: 2 page(s) / 64K

Publishing Venue

IBM

Related People

Breyfogle, FW: AUTHOR [+2]

Abstract

A test design process for verifying a random failure criterion is described. The process offers the test designer unique guidance in choosing the number of permissible failures and acceptance confidence level. Whenever a random failure criterion is high and large sample sizes are acceptable, a test sample size can be determined mathematically given the accept risk, reject risk, and level of uncertainty. More commonly, the failure criterion is low and sample sizes are restrictive. For this situation, the Poisson equation can be used to calculate total test termination time given the number of acceptable failures and acceptance risk. Usually the rejection risk level and rejection sample failure rate is not considered within this approach.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 52% of the total text.

Page 1 of 2

Realistic Random Failure Criterion Certification

A test design process for verifying a random failure criterion is described. The process offers the test designer unique guidance in choosing the number of permissible failures and acceptance confidence level. Whenever a random failure criterion is high and large sample sizes are acceptable, a test sample size can be determined mathematically given the accept risk, reject risk, and level of uncertainty. More commonly, the failure criterion is low and sample sizes are restrictive. For this situation, the Poisson equation can be used to calculate total test termination time given the number of acceptable failures and acceptance risk. Usually the rejection risk level and rejection sample failure rate is not considered within this approach. If neither rejection risk level nor sample rejection failure rate is considered when designing a random failure test certification test, a question can be raised as to the integrity of such a test. It is widely known that the Chi-square distribution can describe a Poisson time terminated test confidence interval. The lower confidence limit is the criterion which is to be verified and takes the form The graph shown in the figure was generated as an example from the above-noted Poisson Equation. It was generated for a given failure rate criterion, acceptance risk
(B), and operating ratio. The lower curve has support failure numbers, while the upper curve has reject failure number. The horizontal axis measures total run time, while the vertical axis is a measure of the "test performance ratio." The test perform...