[SystemSafety] The Patriot Missile Failure (was Re: systemsafety Digest, Vol 34, Issue 5)

[Brent Kimberley]

Perhaps, some systems don't fully monitor / supervise / detect / prevent / control / ... aliasing?
 


     On Saturday, May 9, 2015 7:08 PM, Matthew Squair <mattsquair at gmail.com> wrote:
   

 Reading back my last response, I'd add that a simple counter was probably deemed an 'OK' solution for a mobile application as you would expect it to be shut down every couple of days. Of course that was not how the system was used in practice...
Matthew Squair
MIEAust, CPEngMob: +61 488770655Email; Mattsquair at gmail.comWeb: http://criticaluncertainties.com
On 6 May 2015, at 6:47 pm, Martyn Thomas <martyn at thomas-associates.co.uk> wrote:


 Thanks, Matthew
 
 You show that the RGA is a function of time, but do not explain why "time" has to be the elapsed time since the system was started, rather than (say) gps time, or time since the start of the current radar scan, or ...
 
 Do you know the answer?
 
 Martyn
 
 On 06/05/2015 00:11, Matthew Squair wrote:
  
 Hi Martyn, 
  The time, along with position and velocity, is used in the calculation of a parameter called the range gate which is used in track while scan radar systems to tell the radar where to expect (correlate) a radar return for a tracked object. Clock drift can cause the gate to drift away from the actual radar return's location, once the difference gets past a certain threshold the radar return and track will decorrelate and the radar will loose (or drop) the track.  
  The actual details of how this occurred on the day are a little more complex, if you're interested see the case study at the link for more detail http://wp.me/ax0Kp-2tK.  
 On Tue, May 5, 2015 at 6:56 PM, Martyn Thomas <martyn at thomas-associates.co.uk> wrote:
 
  Why did they need to keep the time, over a period of 100 hours, in order to determine how far away an incoming missile is? The two things appear to me to be unrelated.
 
 Martyn
 
 On 04/05/2015 21:57, Steve Tockey wrote:
  
 
  Can static analysis catch this kind of defect: 
  https://www.ima.umn.edu/~arnold/disasters/patriot.html 
 
  
 _______________________________________________
 The System Safety Mailing List
 systemsafety at TechFak.Uni-Bielefeld.DE
 
 
  
 
 
  -- 
  Matthew Squair  MIEAust CPEng 
  Mob: +61 488770655 Email: MattSquair at gmail.com   Website: www.criticaluncertainties.com 
     
 
 

_______________________________________________
The System Safety Mailing List
systemsafety at TechFak.Uni-Bielefeld.DE


  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.techfak.uni-bielefeld.de/mailman/private/systemsafety/attachments/20150511/ffdec333/attachment-0001.html>