I must not MTBF. MTBF is the mind-killer. MTBF is the little-death that brings total obliteration. I will face my MTBF. I will permit it to pass over me and through me. And when it has gone past I will turn the inner eye to see its path. Where the MTBF has gone there will be nothing. Only Reliability will remain.
- Bene Gesserit Litany Against Fear. (with apologies)
Fred, this was excellent and provided a great laugh. MTBF nearly killed me at the StartUp that Will Not Be Named.
And you can add, “And I will teach everyone I come in contact with why they should no longer MTBF either!”
You really have it in for MTBF, don’t you. ;)
Yes, and this site and my many other ways of getting the word out is part of the effort to eradicate MTBF from common use. I’ve seen way too much confusion, lost profit and delays due to this ‘simple’ metric.
Cheers,
Fred
I’m thinking you’re referring to using MTBF in the early wear out and end of life stages. How do you feel about the assumption of exponential distribution in the middle of the bathtub curve and using MTBF to find failure rate? Assuming that distribution is applicable.
Hi Jim,
I’ve pretty much never have seen a flat part of the bathtub curve. While the change in failure rate over some small amount of time may be considered flat enough – without that evidence it is a risky assumption.
There are always multiple failure mechanisms at play with any product. Even something like drop damage changes over time as user behavior changes from careful new user to careless….for example.
Rather than simplify the calculations unnecessarily focus on the failure mechanisms of most importance using the appropriate modeling and math as an aid.
cheers,
Fred