0% found this document useful (0 votes)
7 views7 pages

MTBF Methods

The document outlines three main methods for calculating Mean Time Between Failures (MTBF): Calculated MTBF, Assessment or Demonstrated MTBF, and Field Observed MTBF. Each method has its own approach, with Calculated MTBF relying on statistical failure rates, Assessment MTBF based on life tests, and Field Observed MTBF using actual failure data from a large population. Environmental conditions significantly impact MTBF values, and the document emphasizes the importance of sample size and testing conditions in obtaining reliable MTBF estimates.

Uploaded by

Dario Alvira
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
7 views7 pages

MTBF Methods

The document outlines three main methods for calculating Mean Time Between Failures (MTBF): Calculated MTBF, Assessment or Demonstrated MTBF, and Field Observed MTBF. Each method has its own approach, with Calculated MTBF relying on statistical failure rates, Assessment MTBF based on life tests, and Field Observed MTBF using actual failure data from a large population. Environmental conditions significantly impact MTBF values, and the document emphasizes the importance of sample size and testing conditions in obtaining reliable MTBF estimates.

Uploaded by

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

August 2015

MTBF Methods

Unrestricted / © Siemens AG 2015. All Rights Reserved. [Link]/ruggedcom


Background

This the key region of interest.

MTBF Valid Only During the Product’s “Useful” Life


Unrestricted / © Siemens AG 2015. All Rights Reserved.
Reliability Methods

Three main types:

1. Calculated MTBF

2. Assessment or Demonstrated MTBF

3. Field observed MTBF

Unrestricted / © Siemens AG 2015. All Rights Reserved.


Calculated MTBF Method

 Prediction method based on statistical failure rates of electronic components, requires a database
 Two popular Methods – MIL-HDBK-217F and BELLCORE/Telcordia (used by Telecom industry)
 Only really useful at the design stage when comparing one design iteration to another
 Often requested by customers

Things to note:

 MIL-217F is more pessimistic (i.e. 2-3x) than Bellcore/Telecordia


So an MTBF value calculated using MIL-217F will be worse e.g. up to 2-3x worse!
 Environmental conditions greatly matter
 Often times vendors will publish calculated MTBF values under “Ground Benign” environmental conditions
at 25°C. This is essentially an air conditioned office compared to “Naval Sheltered” at 55°C there is 2.5-
4x difference in results

Unrestricted / © Siemens AG 2015. All Rights Reserved.


Assessment or demonstrated MTBF Method

 Based on putting a sample number of units on “life test” or reliability demonstration test under greater than
normal stress and then extrapolating for entire population based on failure rate during the test
E.g. Take 100 units and run them at elevated temperatures and electrical stress for a period of
month. Assuming all units survived you would have a point MTBF value of:
100 x 730 hours/month = 73,000 hours.
 Requires a “statistically significant” sample size to be useful (e.g. at least 40 units)

 Hard to distinguish between early failures and infant mortality

Unrestricted / © Siemens AG 2015. All Rights Reserved.


Field observed MTBF Method

 Based on large population size


 Based on actual observed failures in the field
 Final proof, no statistical methods required
 We currently have over 250,000 units in the field over a 11.5 year period.
 1,000,000 hours MTBF = 114 year MTBF
 See Observed MTBF performance chart on following slide

Unrestricted / © Siemens AG 2015. All Rights Reserved.


Thank you for your attention!

[Link]/ruggedcom

Unrestricted / © Siemens AG 2015. All Rights Reserved.

You might also like