TECHNET Archives

March 2000

TechNet@IPC.ORG

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Glenn Pelkey <[log in to unmask]>
Reply To:
Date:
Fri, 24 Mar 2000 14:03:03 PST
Content-Type:
text/plain
Parts/Attachments:
text/plain (92 lines)
Hi Ken,

        I get pretty much the same numbers as the other replies.  Binomial
assumes independent and equally likely.  The Poisson is an approximation of
the Binomial under certain conditions (large sample size (n), small defect
percentage (p), and n x p< or = 7, or so says my thick Stats book).
        Your question is the same as inspection sampling.  Except with
inspection sampling, they also through in the Hypergeometric.  (Here's one of
those terms to through out and impress your friends!)  With reliability,
failures are a function of time.  However, with software, there is no time
dependance (i.e. no aging).

Using Excel,

Binomial:
Sample Size (X) Failures (Y)
----------------------  ----------------
      76                     0
    129              1
    176              2
    221              3
    265              4
    308              5

Poisson:
Sample Size (X) Failures (Y)
----------------------  ----------------
      70                     0
    118              1
    162              2
    203              3
    242              4  Violates Rule, use Binomial
    281              5  Violates Rule, use Binomial

Glenn

Ken Patel <[log in to unmask]> Wrote:
|
| Reliability/Quality Guru,
| I received this question from our engineering department.
|
| Currently, they are seeing 1 system crash out of 10 times. They want to
| improve the crash rate to 1 failure every 30 times with 90% confidence
| level. They want to know how many operation "X" they will run with no more
| than "Y" failures with 90% confidence level which will yield 1 failure every
| 30 times.
|
| I do not know how to solve it as haven't practiced it. Am I providing enough
| information or what information is needed to make a sense out of it? What
| kind of model it will fit, Poisson, Binomial, etc?
|
| Any help will be highly appreciated.
|
| re,
| ken patel
|
|
| ______________________________________________________
| Ken Patel                       Phone:  (408) 490-6804
| 1708 McCarthy Blvd.             Fax:    (408) 490-6859
| Milpitas, CA 95035              Beeper: (888) 769-1808
|
| ##############################################################
| TechNet Mail List provided as a free service by IPC using LISTSERV 1.8c
| ##############################################################
| To subscribe/unsubscribe, send a message to [log in to unmask] with following
text in
| the body:
| To subscribe:   SUBSCRIBE TECHNET <your full name>
| To unsubscribe:   SIGNOFF TECHNET
| ##############################################################
| Please visit IPC web site (http://www.ipc.org/html/forum.htm) for additional
| information.
| If you need assistance - contact Keach Sasamori at [log in to unmask] or
| 847-509-9700 ext.5315
| ##############################################################
|

##############################################################
TechNet Mail List provided as a free service by IPC using LISTSERV 1.8c
##############################################################
To subscribe/unsubscribe, send a message to [log in to unmask] with following text in
the body:
To subscribe:   SUBSCRIBE TECHNET <your full name>
To unsubscribe:   SIGNOFF TECHNET
##############################################################
Please visit IPC web site (http://www.ipc.org/html/forum.htm) for additional
information.
If you need assistance - contact Keach Sasamori at [log in to unmask] or
847-509-9700 ext.5315
##############################################################

ATOM RSS1 RSS2