SRC Forum - Message Replies


Forum: Reliability & Maintainability Questions and Answers

Topic: Reliability & Maintainability Questions and Answers

Topic Posted by: Reliability & Maintainability Forum (src_forum@alionscience.com )
Organization: System Reliability Center
Date Posted: Mon Aug 31 12:47:36 US/Eastern 1998

Back to message list Topics List About this forum
Original Message:

Posted by: Wu Xiangping (7815@notesmail.huawei.com.cn )
Date posted: Tue Aug 3 22:44:36 US/Eastern 1999
Subject: Reliability Prediction
Message:
We all know that the result of reliability prodiction is much higher than the true reliability level.Does anyone know the proportion between them? Another problem is,a new product,has no reliability requirement from customer,how to decide the reliability requirement that the product should achieve? Thanks very much.


Reply:

Subject: Reliability Prediction
Reply Posted by: Clay W Davis
Date Posted: Mon Aug 9 11:25:45 US/Eastern 1999
Message:
I would especially like to know if anyone has had any experience in how accurate the Bellcore predictions are. Agreed, the reliability predictions are conservative but by what factor (1.5, 2, or more). Also, I would presume the more data on the component you can give Bellcore the more accurate the prediction. If any has experience in the Method I Case 3 with electronic circuit boards they would be willing share, I would be grateful.


Reply:

Subject: Reliability Prediction
Reply Posted by: Jon Ilseng (ilseng@austin.apc.slb.com )
Organization: Schlumberger Oilfield Services
Date Posted: Mon Aug 9 15:27:09 US/Eastern 1999
Message:
Regarding the Bellcore Reliability Prediction for Method 1, Case 3. I have done Bellcore Reliability Predictions using Method 1, Case 3 and compared these predictions vs. field data. One example is for a Wireline Interface Telemetry PCB. The Bellcore Reliability Prediction for this PCB using Method 1, Case 3 was: a) first-year MTBF prediction of 9,569 hours (based on a 48-hour burn-in at 50 Degrees C); b) steady-state MTBF prediction of 29,666 hours. The first year this PCB was fielded, the field MTBF was 20,023 hours (compared to the Bellcore MTBF prediction of 9,569 hours). We are only 5 months into the steady-state operating period for this PCB, therefore, it is premature to compare steady-state field MTBF vs. the Bellcore predicted steady-state MTBF. I did calculate a MIL-HDBK-217F,N2 MTBF prediction for this PCB and its calculated MTBF was 11,782 hours. The field MTBF for a 17-month period for this PCB was 17,804 hours. In my 20 years of working in the reliability field, the field MTBFs are usually higher than any calculated MTBFs. By how much, it varies.


Reply:

Subject: Reliability Prediction
Reply Posted by: Bruce Dudley (bdudley@alionscience.com )
Organization: Reliability Analysis Center
Date Posted: Thu Aug 19 9:18:02 US/Eastern 1999
Message:
Not all reliability predictions are much higher than the fielded reliability. In the cases where one is using the older prediction methods that have heavy commercial part quality penalaties and are used in low stress situations, fielded reliability are often greater than predicted. A 1989 study (RADC-TR-89-299 Reliability and Maintainability Operational Parameters Translation) performed by IITRI for the Air Force, developed parameters for translation of predicted values to fielded ones. In general, the ratios are: 0.9 for ground fixed environment, 0.3 for ground mobile environment and 0.4 for aircraft. Specific parameters for equipment types and prediction methods are included in this study.


Reply:

Subject: Reliability Prediction
Reply Posted by: Sumandra Ghosh Chowdhury (sumandra@lucent.com )Lucent Technology
Date Posted: Tue Aug 24 1:04:35 US/Eastern 1999
Message:
My experience always says Predicted MTBF always less than field MTBF. For example for UPS, predicted MTBF as per MIL 217F, was 37000 hours but field MTBF was 50000 hours after 2 years. Initially, field MTBF was in the order of 20000 hours in the first year. Similarly, for a sevocontroller, predicted MTBF was 12000 hours. but field MTBF was 25000 hours. Predicted Field Similarly, Inverter 33000 19000 (Ist year) AC Servo 15000 40000 Converter 18000 21000 CNC 12000 40000 DRO 34000 22000 (Ist year) Volatge Regulator 33000 45000 It is very difficult to assign a factor for same. However, the proportion depends on action based on effective FRACAS, continuous improvement, customer service driven action, more population of products & management focus/attention on than particular product. Reliability Growth occurs as per Duane plot. Thanks Sumandra


Back to message list Topics List About this forum
Reply to this message