[SystemSafety] Discovery of exponential distribution of hardware faults

[Derek M Jones]

Phil,

> Looks like you can order a print copy of that 1952 paper here, although I cannot vouch for how that process might work.

Pre-COVID DTIC would convert microfilm to pdf.
Have not tried since COVID.

A version is also available here
https://www.jstor.org/stable/2280740

> https://www.rand.org/pubs/papers/P183.html
> 
> Davis, Dolores D., An Analysis of Some Failure Data. Santa Monica, CA: RAND Corporation, 1952. https://www.rand.org/ 
> pubs/papers/P183.html. Also available in print form.
> 
> 
> On 11/25/2024 6:31 AM, Derek M Jones wrote:
>> All,
>>
>> The exponential distribution is now the default
>> distribution for hardware failure.  This 1958 paper
>> looks like it might be at the start of the citation
>> tree
>> https://apps.dtic.mil/sti/citations/tr/AD0158605
>>
>> Although this 1952 paper has lots of data and also
>> discusses the exponential distribution.
>> No paper, yet, behind this link
>> https://apps.dtic.mil/sti/citations/tr/AD0422811
>>
>> Are there earlier papers than these?
>>
> 

-- 
Derek M. Jones           Evidence-based software engineering
blog:https://shape-of-code.com