Expected Failure Time Plot: Difference between revisions

From ReliaWiki
Jump to navigation Jump to search
No edit summary
No edit summary
Line 1: Line 1:
<p><span class="fck_mw_template">{{UConstruction}}</span>
<IMG class=FCK__MWTemplate src="http://www.reliawiki.com/extensions/FCKeditor/fckeditor/editor/images/spacer.gif" width=1 height=1 _fckfakelement="true" _fckrealelement="0" _fck_mw_template="true">  
</p>
<h1>Expected Failure Time Plot</h1>
<p>When a reliability life test is planned it is useful to visualize the expected outcome of the experiment. The Expected Failure Time Plot (introduced by ReliaSoft in Weibull++ 8)provides such visual.
</p>
<h2> Background &amp; Calculations  </h2>
<p>Using the cumulative binomial, for a defined sample size, one can compute a rank (Median Rank if at 50% probability) for each ordered failure. As an example and for a sample size of 6 the 5%, 50% and 95% ranks would be as follows:
</p><p><br />
</p>
<table border="1" cellspacing="1" cellpadding="1" width="400" align="center">
<caption> Table 1: 5%, 50% and 95% Ranks for a sample size of 6.&nbsp;
</caption>
<tr>
<th bgcolor="#cccccc" valign="middle" scope="col" align="center"> Order Number
</th><th bgcolor="#cccccc" valign="middle" scope="col" align="center"> 5%
</th><th bgcolor="#cccccc" valign="middle" scope="col" align="center"> 50%
</th><th bgcolor="#cccccc" valign="middle" scope="col" align="center"> 95%
</th></tr>
<tr>
<td valign="middle" align="center"> 1
</td><td valign="middle" align="center"> 0.85%
</td><td valign="middle" align="center"> 10.91%
</td><td valign="middle" align="center"> 39.30%
</td></tr>
<tr>
<td valign="middle" align="center"> 2
</td><td valign="middle" align="center"> 6.29%
</td><td valign="middle" align="center"> 26.45%
</td><td valign="middle" align="center"> 58.18%
</td></tr>
<tr>
<td valign="middle" align="center"> 3
</td><td valign="middle" align="center"> 15.32%
</td><td valign="middle" align="center"> 42.14%
</td><td valign="middle" align="center"> 72.87%
</td></tr>
<tr>
<td valign="middle" align="center"> 4
</td><td valign="middle" align="center"> 27.13%
</td><td valign="middle" align="center"> 57.86%
</td><td valign="middle" align="center"> 84.68%
</td></tr>
<tr>
<td valign="middle" align="center"> 5
</td><td valign="middle" align="center"> 41.82%
</td><td valign="middle" align="center"> 73.55%
</td><td valign="middle" align="center"> 93.71%
</td></tr>
<tr>
<td valign="middle" align="center"> 6
</td><td valign="middle" align="center"> 60.70%
</td><td valign="middle" align="center">
<p>89.09%
</p>
</td><td valign="middle" align="center">
<p>99.15%
</p>
</td></tr></table>
<p><br />
</p><p>Furthermore, consider that for the units to be tested the underlying reliability model assumption is given by a Weibull distribution with <span class="texhtml">&beta; = 2</span>, and <span class="texhtml">&eta; = 100</span> hr.  Then the median time to failure of the first unit on test can be determined by solving the Weibull reliability equation for t, at each probability,
</p><p>or
</p><p><img _fckfakelement="true" _fck_mw_math="R(t)=e^{\big({t \over \eta}\big)^\beta}" src="/images/math/9/b/2/9b21aed609d5cefddaae485bbfbc3a2f.png" />
</p><p>then for 0.85%,
</p><p><br /><img _fckfakelement="true" _fck_mw_math="1-0.0085=e^{\big({t \over 100}\big)^2}" src="/images/math/d/b/e/dbe99885cf4bd0ea65638a820287544a.png" />
</p><p>and so forths as shown in the table below:
</p><p><br />
</p>
<table border="1" cellspacing="1" cellpadding="1" width="400" align="center">


<tr>
= Expected Failure Time Plot =
<th bgcolor="#cccccc" scope="col"> Failure Order Number
 
</th><th bgcolor="#cccccc" scope="col"> Lowest Expected Time-to-failure (hr)
When a reliability life test is planned it is useful to visualize the expected outcome of the experiment. The Expected Failure Time Plot (introduced by ReliaSoft in Weibull++ 8)provides such visual.
</th><th bgcolor="#cccccc" scope="col"> Median Expected Time-to-failure (hr)
 
</th><th bgcolor="#cccccc" scope="col"> Highest Expected Time-to-failure (hr)
== Background &amp; Calculations  ==
</th></tr>
 
<tr>
Using the cumulative binomial, for a defined sample size, one can compute a rank (Median Rank if at 50% probability) for each ordered failure. As an example and for a sample size of 6 the 5%, 50% and 95% ranks would be as follows:
<td valign="middle" align="center"> 1
 
</td><td valign="middle" align="center"> 9.25
<br>
</td><td valign="middle" align="center"> 33.99
 
</td><td valign="middle" align="center"> 70.66
{| border="1" cellspacing="1" cellpadding="1" width="400" align="center"
</td></tr>
|+ Table 1: 5%, 50% and 95% Ranks for a sample size of 6.&nbsp;
<tr>
|-
<td valign="middle" align="center"> 2
! bgcolor="#cccccc" valign="middle" scope="col" align="center" | Order Number  
</td><td valign="middle" align="center"> 25.48
! bgcolor="#cccccc" valign="middle" scope="col" align="center" | 5%
</td><td valign="middle" align="center"> 55.42
! bgcolor="#cccccc" valign="middle" scope="col" align="center" | 50%
</td><td valign="middle" align="center"> 93.37
! bgcolor="#cccccc" valign="middle" scope="col" align="center" | 95%
</td></tr>
|-
<tr>
| valign="middle" align="center" | 1
<td valign="middle" align="center"> 3
| valign="middle" align="center" | 0.85%
</td><td valign="middle" align="center"> 40.77
| valign="middle" align="center" | 10.91%
</td><td valign="middle" align="center"> 73.97
| valign="middle" align="center" | 39.30%
</td><td valign="middle" align="center"> 114.21
|-
</td></tr>
| valign="middle" align="center" | 2
<tr>
| valign="middle" align="center" | 6.29%
<td valign="middle" align="center"> 4
| valign="middle" align="center" | 26.45%
</td><td valign="middle" align="center"> 56.26
| valign="middle" align="center" | 58.18%
</td><td valign="middle" align="center"> 92.96
|-
</td><td valign="middle" align="center"> 136.98
| valign="middle" align="center" | 3
</td></tr>
| valign="middle" align="center" | 15.32%
<tr>
| valign="middle" align="center" | 42.14%
<td valign="middle" align="center"> 5
| valign="middle" align="center" | 72.87%
</td><td valign="middle" align="center"> 73.60
|-
</td><td valign="middle" align="center"> 115.33
| valign="middle" align="center" | 4
</td><td valign="middle" align="center"> 166.34
| valign="middle" align="center" | 27.13%
</td></tr>
| valign="middle" align="center" | 57.86%
<tr>
| valign="middle" align="center" | 84.68%
<td valign="middle" align="center"> 6
|-
</td><td valign="middle" align="center">
| valign="middle" align="center" | 5
<p>96.64
| valign="middle" align="center" | 41.82%
</p>
| valign="middle" align="center" | 73.55%
</td><td valign="middle" align="center"> 148.84
| valign="middle" align="center" | 93.71%
</td><td valign="middle" align="center"> 218.32
|-
</td></tr></table>
| valign="middle" align="center" | 6
<p><br /><br />
| valign="middle" align="center" | 60.70%
</p><p><br />
| valign="middle" align="center" |
</p><p><br />
89.09%
</p><p><br />
 
</p><p><br /><br />
| valign="middle" align="center" |
</p><a _fcknotitle="true" href="Category:Weibull++">Weibull++</a> <a _fcknotitle="true" href="Category:Test_Design">Test_Design</a> <a _fcknotitle="true" href="Category:Special_Tools">Special_Tools</a>
99.15%
 
|}
 
<br>
 
Furthermore, consider that for the units to be tested the underlying reliability model assumption is given by a Weibull distribution with <span class="texhtml">β = 2</span>, and <span class="texhtml">η = 100</span> hr. Then the median time to failure of the first unit on test can be determined by solving the Weibull reliability equation for t, at each probability,
 
or
 
&lt;img _fckfakelement="true" _fck_mw_math="R(t)=e^{\big({t \over \eta}\big)^\beta}" src="/images/math/9/b/2/9b21aed609d5cefddaae485bbfbc3a2f.png" /&gt;
 
then for 0.85%,
 
<br>&lt;img _fckfakelement="true" _fck_mw_math="1-0.0085=e^{\big({t \over 100}\big)^2}" src="/images/math/d/b/e/dbe99885cf4bd0ea65638a820287544a.png" /&gt;
 
and so forths as shown in the table below:
 
<br>
 
{| border="1" cellspacing="1" cellpadding="1" width="400" align="center"
|+ Table 2: Times corresponding to the 5%, 50% and 95% Ranks for a sample size of 6. and assuming Weibull distribution with <span class="texhtml">β = 2</span>, and <span class="texhtml">η = 100</span> hr.;
|-
! bgcolor="#cccccc" scope="col" | Failure Order Number
! bgcolor="#cccccc" scope="col" | Lowest Expected Time-to-failure (hr)  
! bgcolor="#cccccc" scope="col" | Median Expected Time-to-failure (hr)  
! bgcolor="#cccccc" scope="col" | Highest Expected Time-to-failure (hr)
|-
| valign="middle" align="center" | 1  
| valign="middle" align="center" | 9.25  
| valign="middle" align="center" | 33.99  
| valign="middle" align="center" | 70.66
|-
| valign="middle" align="center" | 2  
| valign="middle" align="center" | 25.48  
| valign="middle" align="center" | 55.42  
| valign="middle" align="center" | 93.37
|-
| valign="middle" align="center" | 3  
| valign="middle" align="center" | 40.77  
| valign="middle" align="center" | 73.97  
| valign="middle" align="center" | 114.21
|-
| valign="middle" align="center" | 4  
| valign="middle" align="center" | 56.26  
| valign="middle" align="center" | 92.96  
| valign="middle" align="center" | 136.98
|-
| valign="middle" align="center" | 5  
| valign="middle" align="center" | 73.60  
| valign="middle" align="center" | 115.33  
| valign="middle" align="center" | 166.34
|-
| valign="middle" align="center" | 6  
| valign="middle" align="center" |
96.64
 
| valign="middle" align="center" | 148.84  
| valign="middle" align="center" | 218.32
|}
 
<br><br>
 
<br>
 
<br>
 
<br>
 
<br><br>
 
&lt;a _fcknotitle="true" href="Category:Weibull++"&gt;Weibull++&lt;/a&gt; &lt;a _fcknotitle="true" href="Category:Test_Design"&gt;Test_Design&lt;/a&gt; &lt;a _fcknotitle="true" href="Category:Special_Tools"&gt;Special_Tools&lt;/a&gt;

Revision as of 18:26, 2 March 2011

<IMG class=FCK__MWTemplate src="http://www.reliawiki.com/extensions/FCKeditor/fckeditor/editor/images/spacer.gif" width=1 height=1 _fckfakelement="true" _fckrealelement="0" _fck_mw_template="true">

Expected Failure Time Plot

When a reliability life test is planned it is useful to visualize the expected outcome of the experiment. The Expected Failure Time Plot (introduced by ReliaSoft in Weibull++ 8)provides such visual.

Background & Calculations

Using the cumulative binomial, for a defined sample size, one can compute a rank (Median Rank if at 50% probability) for each ordered failure. As an example and for a sample size of 6 the 5%, 50% and 95% ranks would be as follows:


Table 1: 5%, 50% and 95% Ranks for a sample size of 6. 
Order Number 5% 50% 95%
1 0.85% 10.91% 39.30%
2 6.29% 26.45% 58.18%
3 15.32% 42.14% 72.87%
4 27.13% 57.86% 84.68%
5 41.82% 73.55% 93.71%
6 60.70%

89.09%

99.15%


Furthermore, consider that for the units to be tested the underlying reliability model assumption is given by a Weibull distribution with β = 2, and η = 100 hr. Then the median time to failure of the first unit on test can be determined by solving the Weibull reliability equation for t, at each probability,

or

<img _fckfakelement="true" _fck_mw_math="R(t)=e^{\big({t \over \eta}\big)^\beta}" src="/images/math/9/b/2/9b21aed609d5cefddaae485bbfbc3a2f.png" />

then for 0.85%,


<img _fckfakelement="true" _fck_mw_math="1-0.0085=e^{\big({t \over 100}\big)^2}" src="/images/math/d/b/e/dbe99885cf4bd0ea65638a820287544a.png" />

and so forths as shown in the table below:


Table 2: Times corresponding to the 5%, 50% and 95% Ranks for a sample size of 6. and assuming Weibull distribution with β = 2, and η = 100 hr.;
Failure Order Number Lowest Expected Time-to-failure (hr) Median Expected Time-to-failure (hr) Highest Expected Time-to-failure (hr)
1 9.25 33.99 70.66
2 25.48 55.42 93.37
3 40.77 73.97 114.21
4 56.26 92.96 136.98
5 73.60 115.33 166.34
6

96.64

148.84 218.32








<a _fcknotitle="true" href="Category:Weibull++">Weibull++</a> <a _fcknotitle="true" href="Category:Test_Design">Test_Design</a> <a _fcknotitle="true" href="Category:Special_Tools">Special_Tools</a>