PRISM Forum - Message Replies


Topic: PRISM Questions and Answers
Topic Posted by: SRC (src_forum@alionscience.com )
Organization: SRC
Date Posted: Wed Jan 12 8:33:33 US/Eastern 2000
Topic Description: Welcome to the PRISM forum! Please feel free to post your questions and comments about the PRISM assessment software here.

Back to message list About this forum

Original Message:

Posted by: Jim Weber
Date posted: Wed Apr 27 11:57:30 US/Eastern 2005
Subject: Question about Instantaneous Failure Time
Message:
Could you please explain the meaning of the "Instantaneous Failure Time" factor which can be changed for the Infant Mortality Process Grade? The manual says: "the time when all infant mortality failures have been removed from the system". I understand it that way, that the shorter the "Instantaneous Failure Time" the better the Infant Mortality Process Grade is supposed to be, as (corresponding to the manual) it takes less time to remove infant mortality failures from the system. But in fact the opposite seems to apply. Best Regards, J.W.


Reply:

Subject: Instantaneous Failure Time
Reply Posted by: Norman B. Fuqua (nfuqua@alionscience.com )
Organization: RAC
Date Posted: Tue May 3 13:22:50 US/Eastern 2005
Message:
In researching your question I discovered that the formula for the instantaneous failure rate as recorded in Appendix I of the PRISM Users manual is incorrect. It should be: (1-SSESS)x (t exp -.62/1.77) The Appendix I will be changed to reflect the correct formula.


Reply:

Subject: Response
Reply Posted by: Jim Weber
Date Posted: Mon May 9 6:58:39 US/Eastern 2005
Message:
Thanks for the information. However, I asked about the "Instantaneous Failure Time" and not the Failure Rate... Best Regards, J.W.


Reply:

Subject: Infant Mortality and Instantaneous Failure Time
Reply Posted by: David Dylis (ddylis@alionscience.com )
Date Posted: Mon May 9 10:48:43 US/Eastern 2005
Message:
The degree of impact of infant mortality failure rates on the assembly and system predicted reliability are a function of the temperature screening and vibration screening that are performed prior to the delivery of the product to the customer. PRISM contains areas that address (1) the type of temperature and vibration screening test, (2) the specific conditions (stresses) of the test, (3) the detection efficiency associated with the test, (4) the estimated percent of potential infant mortality failures resulting from temperature cycling and vibration for each screening test type option, and (5) the instantaneous failure time (the time when all infant mortality failures have been removed from the system) baseline (in months).

When the default option is accepted (no temperature or vibration screening), the only fields accessible to the user are "Percent Failures from Temperature Cycling" (defaulted to 80%), "Percent Failures from Vibration" (defaulted to 20%), and the instantaneous failure time (defaulted to 5 months). The user can change these defaults by incrementing the value using the up/down counter next to the field (where applicable), or by directly typing in the desired value in the field. Note the sum of percentages must be 100%. Therefore, changing the value of one field will automatically adjust the value of the other. For example, if the user changes the value of the "Percent Failures from Temperature Cycling" field to 67%, the value of the "Percent Failures from Vibration" will automatically change to 33%, and vice versa. When these values change, the pi-factor value is automatically re-calculated.

The infant mortality failure rate is a decreasing function over time. When a user only modifies the “instantaneous failure rate” value in PRISM without modifying other default parameter values used in the infant mortality Pi factor calculation he is only moving the instantaneous failure rate point on the infant mortality failure rate curve. Therefore, if the user chooses a smaller instantaneous failure rate and keeps all other PRISM infant mortality default parameter values constant, more infant mortality failures are injected into the field increasing the predicted failure rate.

Based on the PRISM defaults, our data has shown that the instantaneous failure rate time is 5 months for a typical system, however, we allow the user to change all default values if he has data to substantiate a different figure.


Reply:

Subject: Instantaneous Failure Time
Reply Posted by: Barry Ives Lockheed Martin Systems Integration - Owego, NY
Date Posted: Thu Jun 16 8:29:44 US/Eastern 2005
Message:
After reading all the responses on this subject, I have the following question: If, as David states, “t”, the instantaneous failure time, represents a point on the infant mortality curve, then it does not represent “the time when all infant mortality failures have been removed”. Does it represent a time when SSESS % of failures have been removed, or when 1-SSESS % of failures have been removed, or something else? Thank-you. Barry


Reply:

Subject: Instantaneous Failure Time
Reply Posted by: David Dylis Reliaiblity Analysis Center (RAC)
Date Posted: Thu Jun 16 10:14:03 US/Eastern 2005
Message:
You are correct to identify that time (t) does not represent the time when all infant mortality failures have been removed from the system. When time(t) is reduced, less infant mortality failures are removed from the fielded system, resulting in an increased field failure rate.


Back to message list About this forum

Reply to this message