Re: [SystemSafety] Software reliability (or whatever you would prefer to call it)

From: Mike Ellims < >
Date: Tue, 10 Mar 2015 16:31:51 -0000


I suspect that you may be wrong.

Going to the bookshelf I found my copy of, Practical Reliability Engineering (4th Ed) 2002. One of the standard texts on reliability engineering (and quite a good read for a text book).

In chapter 1 there is a section "Why do engineering items fail?" In summary;

  1. The design may be inherently incapable
  2. The item may be overstressed in some way.
  3. Failures might be caused by variation
  4. Failures can be caused by wearout.
  5. Failures can be caused by other time dependant mechanisms.
  6. Failures can be caused by sneaks.
  7. Failures can be caused by errors.
  8. There are many other potential caused of errors!

Number 7 further states that "Failures can be caused by errors, such as incorrect specifications, designs or software coding, by faulty assembly or test, by inadequate maintenance, or by incorrect use. The actual failure mechanisms that result might include most of the list above".

Further Chapter 10 is titled "Software Reliability". The chapter skims several models for software failure, namely...

1. Poisson model
2. Musa model
3. Jelinski-Moranda and Schick-Wolverton models
4. Littlewood models.

The first edition came out in 1981 and a fifth edition is out (2012) The blurb on Amazon describes it as... "the essential reliability textbook" so I suspect it may have some authority.

Just saying ;-)

-----Original Message-----
From: systemsafety-bounces_at_xxxxxx [mailto:systemsafety-bounces_at_xxxxxx Michael Holloway
Sent: 10 March 2015 13:06
To: systemsafety_at_xxxxxx Subject: Re: [SystemSafety] Software reliability (or whatever you would prefer to call it)

I believe that if one looks at the history of reliability theory, one would discover that I'm not the one doing the redefining. But I may be wrong.

(standard disclaimer)

On 3/10/15 8:16 AM, Martyn Thomas wrote:
> My train service is unreliable - not because it degrades physically
> (though it does) but because the drivers don't show up when needed ...
> (Maybe they have degraded physically too, but I blame late nights
> watching football).
>
> Let's not try to redefine "reliability". Let's characterise the
> properties that we are discussing and choose some names that we agree
> to use for them (without overloading terms that have other, accepted
meanings).
>
> Martyn
>
> On 10/03/2015 10:34, C. Michael Holloway wrote:
>> I can't speak for Nick, but I object to the use of the term
>> "reliability" being applied to anything other than failures (using
>> the term loosely) resulting from physical degradation over time.
> _______________________________________________
> The System Safety Mailing List
> systemsafety_at_xxxxxx >



The System Safety Mailing List
systemsafety_at_xxxxxx
---
This email has been checked for viruses by Avast antivirus software.
http://www.avast.com

_______________________________________________
The System Safety Mailing List
systemsafety_at_xxxxxx
Received on Tue Mar 10 2015 - 17:32:14 CET

This archive was generated by hypermail 2.3.0 : Wed Feb 20 2019 - 02:17:07 CET