Escolar Documentos
Profissional Documentos
Cultura Documentos
ReliaSofts
ALTA 6
No part of this training guide may be reproduced or transmitted in any form or by any means, for
any purpose, without the expressed written consent of ReliaSoft Corporation, Tucson, AZ, USA.
ReliaSoft and ALTA are trademarks of ReliaSoft Corporation.
ReliaSoft Corporation
ReliaSoft Plaza
115 South Sherwood Village Drive
Tucson, AZ 85710
USA
+1.520.886.0366
Support@ReliaSoft.com
http://ALTA.ReliaSoft.com
Contents
2 Features Summary...................................................................................3
2.1
2.2
2.3
2.4
Contents
ALTA 6
Example 14 ............................................................................................................................65
Example 15 ............................................................................................................................72
Example 16 ............................................................................................................................77
Example 17 ............................................................................................................................85
Example 18 ............................................................................................................................93
Example 19 ..........................................................................................................................102
Example 20 ..........................................................................................................................108
ii
http://ALTA.ReliaSoft.com
Introduction
ALTA 6
http://ALTA.ReliaSoft.com
2 Features Summary
Features Summary
ALTA 6
ALTA includes all the tools and options you will need for accelerated life testing analyses. You can easily
analyze data from complete or suspended (right censored) censoring schemes and enter the data individually
or in groups of observations.
Stress Profile Explorer, which allows you to analyze data from accelerated tests where the stress
profile varies with time. Stress Profile Explorer is only available in ALTA 6 PRO.
Degradation Analysis, which allows you to extrapolate the failure times of a product based on its
performance over a period of time under accelerated stress conditions. You can transfer the extrapolated
failure times into the current Data Folio for further analysis.
Distribution Wizard, which determines the best distribution for your current data set.
Likelihood Ratio Test, which allows you to validate the consistency of the shape parameter, beta or
standard deviation, across multiple stress levels for the currently calculated data set.
You can view or modify the use stress values for each stress type at any point in the analysis using the
Use Stress Level window.
You can specify which stress columns are considered when the parameters are calculated in the Select
Stress Level window.
You can select a stress transformation to be used for each of the selected stress types when the
parameters are calculated with the General Log-Linear model in the Stress Relation window. The
General Log-Linear model and Stress Relation window are only available in ALTA 6 PRO.
4 http://ALTA.ReliaSoft.com
Features Summary
ALTA 6
Side-by-Side Plots tool, which allows you to view up to nine different plots at a time in a single window.
Three new residuals plot types: Standardized, Cox-Snell and Standardized vs. Fitted Values.
Option to shade the upper and lower confidence bounds area on probability plots.
RS Draw, an integrated metafile graphics editor, allows you to take complete graphical control of all plots
and graphics you create and to annotate and enhance any aspect of the plot. In Version 6, RS Draw has an
improved and easier-to-use interface and offers a new custom arrow tool.
ALTA's general spreadsheets provide another flexible option for custom analyses. The spreadsheets can be
used just like an Excel worksheet, with complete in-cell formula support, cell references and over 140 builtin functions. ReliaSoft's new advanced Function Wizard can be used with the General Spreadsheet and
Report Work Center to generate a variety of calculated results. The Function Wizard includes multiple
functions that return additional results from any referenced data set. The Function Wizard in ALTA 6 is now
capable of nesting these functions, which allows you to combine multiple functions to create a composite
function that meets your particular requirements. You can also use the Function Wizard to insert plot
graphics into your reports and analyses.
http://ALTA.ReliaSoft.com 5
Features Summary
ALTA 6
and location of any of the toolbars and customize the icons displayed in each toolbar. In addition, you can set
which tools and options you want displayed in the Data Folio's Control Panel.
6 http://ALTA.ReliaSoft.com
3 First Steps
First Steps
ALTA 6
http://ALTA.ReliaSoft.com
First Steps
ALTA 6
Create a new Data Folio by selecting New from the File menu or by clicking the New icon.
The Data Type Expert window will appear like the one shown next. (If the Data Type Wizard is
displayed instead, click the Data Type Expert button in the bottom left corner.) This window helps you
select the type of Data Entry Spreadsheet you want to create in your Folio.
The Data Type Expert allows you to select the type of Data Entry Spreadsheet that will be
appropriate for your data and the stress type(s) you will be using. The data types are broken
down into two groups: Non-Grouped Data Entry and Grouped Data Entry. In the Select Each
Stress To Use area, you can select and define the stress(es) that will appear within the created
Data Entry Spreadsheet. However, if you are unsure of the data type to be entered, you can
click the Data Type Wizard button in the bottom left corner of the window. The Data Type
Wizard allows you to move step-by-step through a series of questions designed to determine
the type of Data Entry Spreadsheet that will be appropriate for your data.
From the Non-Grouped Data Entry options, select Times-to-Failure (No Censoring). Select
Temperature K under Select Each Stress To Use. The text of the button will be red to indicate that it is
selected. To undo a selection, click the stress button again and the button text will turn black. Note that
if you select Temperature as the stress type, but then enter voltage values within the Data Folio, the
calculations will still be performed correctly. These stress buttons simply allow you to set up the
appearance of the Data Folio (e.g. the column headings, etc.). They do not have any effect on the data
being entered or the calculations.
http://ALTA.ReliaSoft.com
First Steps
ALTA 6
In the Data Type Expert, right-click the Temperature K button to open the Stress Properties window.
The Stress Properties window allows you to specifically define the stress. For this example, accept the
default settings for the Stress Name (Temperature), Stress Units (K) and Stress Type (Thermal). The
Stress Name and Stress Units will be used as the heading for the stress column within the data sheet. For
the Use Stress, enter 356, as shown next.1
Except for the use stress level, the other stress properties do not affect calculations.
Click OK to accept the current settings and close the Stress Properties window. Click OK in the Data
Type Expert. A new Data Folio with the Data Entry Spreadsheet you defined will be created.
The Data Entry Spreadsheet is where all the data are entered in the Folio. Notice that the heading of the
second column displays the stress name and stress units, which were defined in the Stress Properties
window.2
Enter the times-to-failure points in the Time to Failure column and the corresponding stress levels in
the Temperature K column, as shown next.
Note that the times-to-failure do not need to be sorted; the application will do this for you automatically.
The next step is to select a life-stress relationship and an underlying life distribution to calculate the
parameters. We assume that you have not changed any of the default settings. The default life-stress
relationship is Arrhenius. You can verify this by noticing that Arrhenius is selected from the Life-Stress
1.
The use stress level can be changed at any time within the Data Folio by clicking the Use Stress button on the Control
Panel.
2.
The column headings can be edited in the Data Entry Spreadsheet by double-clicking the heading and typing a new
name in the window that appears.
10
http://ALTA.ReliaSoft.com
First Steps
ALTA 6
Relationship drop-down menu on the top of the Main page of the Data Folio Control Panel, as shown
next.
The default underlying life distribution is the Weibull distribution. Also, all calculations in ALTA 6 use
MLE (Maximum Likelihood Estimation) as the parameter estimation method. You can check this by
looking at the Information area at the bottom of the Main page of the Data Folio Control Panel, as
shown next.
You can also view the selected distribution by clicking the Set Analysis page index tab and noticing that
Weibull is selected. Now return to the Main page of the Data Folio Control Panel by clicking the Main
page index tab.
Once the life-stress relationship and distribution have been selected, you can calculate the parameters
by clicking the Calculate icon,
or by selecting Calculate Parameters from the Data menu. The Data Entry Spreadsheet with its
parameters calculated is shown next. The results will appear in the Results Area of the Control Panel,
which is located just below the life-stress relationship options.
or by selecting Plot Probability from the Data menu. If you click the Plot icon before the parameters
have been calculated, the application will automatically calculate the parameters and then plot the data.
http://ALTA.ReliaSoft.com
11
First Steps
ALTA 6
The Use Level Weibull Probability plot should appear in the Plot Sheet, as shown next.
A legend that presents information about the data being plotted is displayed on the right side of the plot. This
includes the life-stress relationship and distribution, the stress level and the number of failures/suspensions
in the data set. The values for the parameters for the data set are located in the bottom left corner. The legend
and results can be disabled by de-selecting Show Results and Show Legends from the Plot Options menu.
By now, with just a few mouse clicks, you have calculated the parameters of the ArrheniusWeibull model and plotted the results. In addition, ALTA 6 gives you the flexibility to adjust and
customize all aspects of the plot from the Plot Setup. Plot Setup is presented in detail in the User's
Guide. For this example, we will not discuss changing the default setup.
At this time, save the data file as 1stStep.ra6. To do this, click the Save icon,
12
http://ALTA.ReliaSoft.com
First Steps
ALTA 6
In the window that appears, select to save the file in the ALTA6/Training Guide folder and type 1stStep
for the filename, as shown below. You do not have to include .ra6 after the filename because ALTA will
do this for you. Click OK.
After saving the file, close the Data Folio by clicking the close button (X) in the upper right corner of
the Data Folio or by selecting Close from the File menu. You should now be looking at the MDI with no
Folios open.
http://ALTA.ReliaSoft.com
13
The following examples can be performed using ALTA 6 Standard and ALTA 6 PRO. Examples that
are specific to ALTA 6 PRO can be found in Chapter 5.
4.1 Example 1
Since the life of a new product under normal operating conditions is expected to be more than 15,000
hours, testing under these conditions is not time-wise feasible. For this reason, it was decided to run
an accelerated test for this product. The operating temperature for this product is 323K (50C) and
temperature is the only acceleration variable. A table of the data obtained from the test for three
different accelerated stress levels is given below.
Stress Level, K
Times-to-Failure, hr
393K
3850
4340
4760
5320
5740
6160
6580
7140
7980
8960
408K
3300
3720
4080
4560
4920
5280
5640
6120
6840
7680
423K
2750
3100
3400
3800
4100
4400
4700
5100
5700
6400
ALTA 6
Do the following:
Create the Weibull Probability and Use Level Weibull Probability plots.
Solution
The file for this example is saved in the "Training Guide" folder and is named "Example1.ra6."
The Data Type Expert will appear as shown next. If the Data Type Wizard opens instead, click the Data
Type Expert button in the bottom left corner of the window to open the Data Type Expert.
Select Times-to-Failure (No Censoring) under the Non-Grouped Data Entry options. Click the
Temperature K button to specify temperature as the stress type for this data set.
Right-click the Temperature K button to open the Stress Properties window, which allows you to
specify the stress properties. For this example, accept the default settings for the Stress Name
(Temperature), Stress Units (K) and Stress Type (Thermal). For the Use Stress, type 323 as shown next.
Click OK to accept the current properties and return to the Data Type Expert.
16
http://ALTA.ReliaSoft.com
ALTA 6
In the Data Type Expert, click OK. A Data Folio will open with the appropriate Data Entry Spreadsheet.
Enter the data into the Data Entry Spreadsheet. Enter the times-to-failure values in the Time to Failure
column and enter the temperatures in the Temperature K column, as shown next. (Please note that the
figure below only displays 15 rows of data. Be sure to enter all of the data for this example as given in
the table on page 15. Your Data Entry Spreadsheet will contain 30 rows of data.)
The Arrhenius life-stress relationship and the Weibull distribution will be used to calculate the
parameters. Make sure that Arrhenius is selected. Weibull should be displayed in the top left box of the
Information area. If Weibull is not displayed, switch to the Set Analysis page and select Weibull.
If you switched to the Set Analysis page of the Folio Control Panel to change the distribution, return to
the Main page by clicking the Main page index tab.
http://ALTA.ReliaSoft.com
17
ALTA 6
or by selecting Calculate Parameters from the Data menu. The results will appear in the Results area,
as shown below.
or by selecting Plot Probability from the Data menu. The Use Level Weibull Probability plot for this
data is shown next.
The Use Level Weibull Probability plot is based on the value of the use stress level that you entered
when you created the Data Entry Spreadsheet, 323K. Changing the value of the use stress level will
change the location of the probability plot.
18
http://ALTA.ReliaSoft.com
ALTA 6
You can also plot the Weibull Probability plot by selecting Probability-Weibull from the Special Plot
Type drop-down menu on the Plot Sheet Control Panel.
Notice that each accelerated stress level is plotted on the probability plot along with the use stress level.
You can select which stress levels appear on the plot using the Plot Stress Levels window, which can be
opened by selecting Plot Stress Levels from the Plot Options menu. The Plot Stress Levels window is
shown next.
Click the stress levels from the current Data Entry Spreadsheet that you want to appear on the plot. You
can click Select All Available to select all the available stress levels or click Clear All Available to deselect all the selected stress levels.
http://ALTA.ReliaSoft.com
19
Now save this Folio as Example1.ra6 and then close the Folio.
ALTA 6
4.2 Example 2
What is the unreliability of the units in Example 1 for a mission duration of 9,000 hours, starting the mission
at T = 0 with an operating stress level of 323K?
Solution
There are two methods of solution for this problem. The first and more laborious method is to extract the
information directly from the plot. This can be done using ReliaSoft Draw (RS Draw). The second method is
to use ALTA's Quick Calculation Pad (QCP) to obtain the exact result. Both methods are presented next.
or by selecting Open from the File menu. Select Example1.ra6, as shown below.
20
http://ALTA.ReliaSoft.com
ALTA 6
The Folio you created in Example 1 will appear like the one shown next.
Click the Plot of Data 1 page index tab to display the plot of the data. Make sure that the Use Level
Weibull Probability plot is being displayed. For this example, the unreliability for a mission of 9,000 hr
is required. However, the minimum value for the X-axis in the use level probability plot is 10,000 hr.
You can change the scaling of the plot using the X and Y scaling boxes on the Plot Sheet Control Panel.
First, de-select the Automatic Scaling option below the scaling boxes to disable the automatic scaling.
This will activate the X and Y scaling boxes. Next, enter the new value within the appropriate input box.
In this case, change the lower limit of the X-axis from 10,000 to 1,000 as shown next.
After typing the new value, press Enter if the Auto Refresh option is selected. If the Auto Refresh
option is not selected, click the Refresh icon.
After the plot has been refreshed, click the RS Draw icon,
or select ReliaSoft Draw from the Tools menu. RS Draw can automatically track the position of the
mouse cursor and translate the coordinates for you. Obtain the unreliability at T = 9,000 hr by placing
the cursor at the intersection of the plotted line and T = 9,000. The position of the cursor is indicated by
the Position Indicator located in the lower right corner of the RS Draw window. The x-coordinate (time)
is displayed on the left and the y-coordinate (unreliability) is displayed on the right. When the xcoordinate reads approximately 9,000, read off the value of the y-coordinate. You may not be able to
obtain the value of the unreliability at exactly 9,000 hours. This is one of the disadvantages of reading
http://ALTA.ReliaSoft.com
21
ALTA 6
the value off the plot. The position of the cursor at the intersection of the plotted line and T 9000 is
shown next.
The value of the y-coordinate from the Position Indicator, the unreliability Q(T), is such that Q(T)
Q ( T ) 4.07 %. Therefore, a good estimate of the probability of failure (unreliability) at 9,000 hours is
4.07%.
Close RS Draw. Return to the Data Entry Spreadsheet by clicking the Data 1 page index tab.
The second method involves the use of the Quick Calculation Pad (QCP). Open the Quick Calculation
Pad by clicking its icon,
or by selecting Quick Calculation Pad from the Tools menu. When the QCP appears, do the following:
On the Basic Calculations page, select Results as Probability of Failure under Results Option.
22
http://ALTA.ReliaSoft.com
ALTA 6
Under Required Input From User, enter 323 for Temperature and 9,000 for the Mission End Time.
Click Calculate to obtain the results shown next.
The QCP returned a result of Q(T) = 4.16%. This is similar to the result found using the plot. However,
the QCP is obviously more accurate and easier to use.
Close the QCP by clicking Exit but leave the Example1.ra6 Data Folio open.
4.3 Example 3
Using Example1.ra6, what is the reliability for a mission duration of T = 3,000 hr, starting the mission at T =
6,000 hr at a use stress level of 323K? Also, what is the warranty time for a reliability of 99% at the same
stress level?
Solution
or by selecting Quick Calculation Pad from the Tools menu. Do the following:
On the Basic Calculations page, select Results as Reliability under Results Option.
Select Conditional Calculations under Options for Calculations.
Under Required Input From User, enter 323 for the Temperature.
Enter the Mission Start Time = 6000 and the Mission Additional Time = 3000.
http://ALTA.ReliaSoft.com
23
ALTA 6
The conditional reliability starting the mission at 6,000 hr and for a 3,000 hr duration is equal to
96.56%. This satisfies the first part of the example.
The second part of the example requires the determination of the warranty time for a reliability of 99%.
There are multiple ways to obtain this information using ALTA 6, using calculation or graphical
methods. They include:
Use the Warranty (Time) Information calculation in the Quick Calculation Pad to calculate the
value. This calculation method is the fastest and most accurate method and it is described below
(first).
Use the Life vs. Stress plot and the Specify Life Lines option. This graphical method is described
below (second).
Use the Warranty Information function in the Function Wizard. The Function Wizard can be
accessed with a General Spreadsheet or the Report Work Center and Section 4.9 of this training
guide provides an example of the use of this utility. This is similar to the QCP calculation method
and it is not described here.
Use the Use Level Weibull Probability plot and RS Draw. This is similar to the procedure
described in Section 4.2 and it is not described again here.
To use the QCP calculation method to determine the warranty time for 99% reliability, do the following:
On the Basic Calculations page, select Warranty (Time) Information under Options for
Calculations.
Enter the Required Reliability = 99.
24
http://ALTA.ReliaSoft.com
ALTA 6
The time required for a reliability of 99% is equal to 6431.1 hr. Close the QCP by clicking Exit.
To use the Life vs. Stress plot graphical method to determine the warranty time for 99% reliability, do
the following:
Click the Plot of Data 1 page index tab to display the plot of the data. The Use Level Weibull
Probability plot will probably be displayed. Select Automatic Scaling and switch to the Life vs.
Stress plot by selecting Life vs Stress under Special Plot Type.1
For the Life vs. Stress Plot, the scale parameter (Eta), mean life, scale parameter and imposed pdf
will appear on the plot by default. You can also add up to ten additional lines corresponding to ten
different percentiles (unreliabilities) by following the steps described next.
1.
If Auto Refresh is not selected, you must click the Refresh icon for the changes to take effect on the plot.
http://ALTA.ReliaSoft.com
25
ALTA 6
Select Show Life Char. Lines and then Specify Life Lines from the Plot Options menu. The
Specify Life Lines window will appear. In this window you can enter up to ten percentile values
to be plotted or delete existing values. You can also specify whether to display confidence bounds
for each unreliability value. Click the box under Show CB to indicate that you would like
confidence bounds to be plotted for the corresponding line. The Show Confidence Bounds
option under the Plot Options menu must also be selected for the confidence bounds to be
plotted.
For this example, enter 99 for the unreliability value for the first line and 1 for the unreliability for
the second line. Do not select to show the confidence bounds for the lines (i.e. do not select the
Show CB box.)
26
http://ALTA.ReliaSoft.com
ALTA 6
The Life vs. Stress plot will look like the one shown next.2
Life vs Stress
1.00E+5
Arrh/Weib
Data 1
Life
Eta
Mean Life
99
1
323
F=30 | S=0
10000.00
User's Name
Company
6/3/02 11:34
1000.00
300.00
328.00
356.00
384.00
412.00
440.00
Temperature
The bottom line corresponds to 1% unreliability, the second line is the mean life, the third line is the
scale parameter (Eta) and the top line corresponds to 99% unreliability.
De-select Automatic Scaling and change the scaling to the values shown next. If Auto Refresh is
selected, click Enter to refresh the plot to reflect the changes made to the scaling. If Auto
Refresh is not selected, then click the Refresh icon.
2.
If Auto Refresh is not selected, you must click the Refresh icon for the changes to take effect on the plot.
http://ALTA.ReliaSoft.com
27
ALTA 6
The position of the cursor at the intersection of the plotted line and stress = 323 is shown next.
The value of the y-coordinate from the Position Indicator is equal to 6453.60. Therefore, a good
estimate of the warranty time for a 99% reliability is 6453 hr. This is similar to the value obtained using
the QCP. However, the QCP is obviously more accurate and easier to use.
Save and close the Folio.
28
http://ALTA.ReliaSoft.com
ALTA 6
4.4 Example 4
Twelve electronic devices were put in a continuous accelerated test. The accelerated stresses were
temperature and voltage, with use level conditions of 328K and 2V. The data obtained are shown in the table
below.
Time,
hr
620
632
658
822
380
416
460
596
216
246
332
400
Temperature,
K
348
348
348
348
348
348
348
348
378
378
378
378
Voltage,
V
3
3
3
3
5
5
5
5
3
3
3
3
Do the following:
Obtain the Life vs. Stress plot for temperature (Life vs. Temperature).
Obtain the Life vs. Stress plot for voltage (Life vs. Voltage).
Solution
The file for this example is saved in the "Training Guide" folder and is named "Example4.ra6."
http://ALTA.ReliaSoft.com
29
ALTA 6
The Data Type Expert will appear as shown next. If the Data Type Wizard opens instead, click the Data
Type Expert button in the bottom left corner of the window to open the Data Type Expert.
Select Times-to-Failure (No Censoring) from the Non-Grouped Data Entry options and select
Temperature and Voltage for stress types, as shown above.
Right-click the Temperature button to open the Stress Properties window, which allows you to define
the stress properties. Accept the default settings for Stress Name (Temperature), Stress Units (K) and
Stress Type (Thermal). Type 328 for the Use Stress, as shown next.
Right-click the Voltage button to open the Stress Properties window. These properties apply to the
voltage stress. Accept the default settings for Stress Name (Voltage), Stress Units (V) and Stress Type
(Electrical). Type 2 for the Use Stress, as shown next.
Click OK to return to the Data Type Expert. Click OK to generate the appropriate Data Entry
Spreadsheet.
30
http://ALTA.ReliaSoft.com
ALTA 6
Enter the data into the Data Entry Spreadsheet, as shown next.
Select Temperature-NonThermal as the life-stress relationship. Make sure that Weibull is selected as
the distribution, or select it as explained in the previous examples.
or by selecting Calculate Parameters from the Data menu. The calculated parameters are shown next.
http://ALTA.ReliaSoft.com
31
ALTA 6
The Use Level Weibull Probability plot will appear. Select Probability-Weibull as the Special Plot
Type and refresh the plot.
Probability Weibull
99.00
T-NT/Weib
Data 1
328 / 2
90.00
348 / 3
F=4 | S=0
348 / 5
F=4 | S=0
378 / 3
F=4 | S=0
Unreliability
50.00
10.00
5.00
User's Name
Company
6/3/02 15:56
1.00
100.00
1000.00
10000.00
Time
Four lines will appear on the probability plot, including three lines for the three combinations of
stresses: (348K, 3V), (348K, 5V) and (378K, 3V); and one line for the use stress level (328K, 2V).
Next, use the Quick Calculation Pad to obtain the warranty time for a reliability of 95%. The QCP can
be accessed while a Plot Sheet is in view by selecting Quick Calculation Pad from the Tools menu.
When the QCP appears, do the following:
On the Basic Calculations page, select Warranty (Time) Information from the Options for
Calculations.
Under Required Input From User, a drop-down menu will appear that contains both stress types,
as shown next. Select a stress type from the menu and enter the corresponding values in the input
box: 328 for the Temperature and 2 for the Voltage. Enter .95 for the Required Reliability.
32
http://ALTA.ReliaSoft.com
ALTA 6
The warranty time for a reliability of 95% is 1,003.6465 hr. Close the QCP and return to the Plot Sheet
by clicking Exit.
The Life vs. Stress plot can be obtained by selecting Life vs Stress as the Special Plot Type in the Plot
Sheet Control Panel. The Life vs. Stress plot will appear, as shown next.
Life vs Stress
10000.00
T-NT/Weib
Data 1
Eta
Mean Life
328 / 2
F=12 | S=0
Life
1000.00
User's Name
Company
6/3/02 16:00
100.00
300.00
318.00
336.00
354.00
372.00
390.00
Temperature
http://ALTA.ReliaSoft.com
33
ALTA 6
Open the Use Stress Level window by clicking the Use Stress button on the control panel.
An option labeled Vary appears to the right of each stress type, which allows you to specify for one of
the stresses to be varied while the other stresses are held constant at the use stress level (i.e. if Vary is
selected for one stress, then the other stresses will be held constant.)
In the previous plot, the Temperature stress is held constant, which means that the temperature was
plotted and the Voltage stress remained constant at 2V.
To obtain a Voltage vs. Stress plot, select the Vary option for the Voltage stress, as shown next.
34
http://ALTA.ReliaSoft.com
ALTA 6
From the Plot Sheet, click Refresh and the Life vs. Voltage plot for a constant temperature of 328K will
appear as shown next.
Life vs Stress
10000.00
T-NT/Weib
Data 1
Life
Eta
Mean Life
328 / 2
F=12 | S=0
1000.00
User's Name
Company
6/3/02 16:04
100.00
1.00
10.00
Voltage
http://ALTA.ReliaSoft.com
35
ALTA 6
4.5 Example 5
A tensile component of a landing gear was put through an accelerated reliability test to determine whether
the return life goal would be achieved under the designed-in load. Fifteen units, N = 15, were tested at three
different shock loads. The component was designed for a peak shock load of 40 kips with an estimated
return of 10% of the population by 10,000 landings. Using the Inverse Power Law-Lognormal model,
determine whether the designed life requirement was met.
Number of Units
1
3
1
3
1
1
1
1
1
1
1
Failure, in Landings
(simulated)
1176
1512
3528
624
816
1296
204
228
252
300
324
Load, kips
73
73
73
98
98
98
123
123
123
123
123
Solution
The file for this example is saved in the "Training Guide" folder and is named "Example5.ra6."
To analyze grouped data, simply create a new Data Folio by clicking the New icon,
or by selecting New from the File menu. The Data Type Expert will appear.
36
Select Times-to-Failure (No Censoring) from the Grouped Data Entry options. Select Mechanical
as the stress type.
http://ALTA.ReliaSoft.com
ALTA 6
Right-click the Mechanical button to open the Stress Properties window. Change the Stress Name to
Load and the Use Stress to 40. Then click OK to accept the settings and to return to the Data Type
Expert.
Notice that the button text in the Data Type Expert now reads Load to reflect the new Stress Name.
Finally, click OK to create the appropriate Data Entry Spreadsheet. Enter the data as shown next. Note
the titles of the column headings.
You can change the heading of the second column, currently Time to Failure, by double-clicking the
heading. A window will appear, which allows you to change the heading. Type Simulated Landings
and click OK. The row headings can be changed in the same manner, if desired.
Select Inverse Power Law as the life-stress relationship and Lognormal as the underlying life
distribution.
Calculate the parameters and open the QCP by clicking its icon,
http://ALTA.ReliaSoft.com
37
ALTA 6
Select Warranty (Time) Information. Enter 40 for the Load and 90% for the Required Reliability.
Click Calculate to obtain the results shown next.
The life is estimated to be 10,545 landings, which meets the designed specifications. We will continue
this example and perform further estimations using the QCP in the next section. Keep the file open and
proceed to Example 6.
38
http://ALTA.ReliaSoft.com
ALTA 6
4.6 Example 6
In the previous example, it was found that the 10,000 landings life criteria for a 90% reliability was met.
However, the estimate of 10,545 landings is very close to the requirement. Additionally, this estimate was
obtained at the 50% confidence level. In other words, 50% of the time life will be greater than 10,545
landings and 50% of the time life will be less. Repeat the calculation of the previous example and this time
include a 90% lower 1-sided confidence bound on the estimation of the Warranty (Time) Information.
The QCP should still be open from the previous example. Warranty (Time) should still be selected with
a required reliability of 90% and a stress level of 40.
Next, go to the Confidence Bounds page by clicking the Confidence Bounds tab. Select Show
Confidence Bounds. More options will appear on the page. Enter 90 in the Confidence Level box.
Select Show Lower One-Sided and click Calculate to obtain the results shown next.
The 90% lower limit was estimated to be 6729.9599 landings, which is below the designed life of
10,000 landings.
http://ALTA.ReliaSoft.com
39
ALTA 6
4.7 Example 7
The following example involves using a data set that contains suspended (right-censored) data.
A certain electronic unit was put through an accelerated test. It was decided to perform the test at four
different temperature levels; 423K, 443K, 463K and 493K. The use stress level is 393K. The data set
obtained from the test is shown next.
Stress Level, K
Times-to-Failure, hr
Stress Level, K
Suspension Time, hr
423K
-
443K
1764
2772
3446
3542
3780
4680
5196
463K
408
1120
1344
1402
1440
-
493K
408
456
504
504
504
-
423K
6064
6064
6064
6064
6064
6064
6064
6064
6064
6064
443K
5448
5448
5448
-
463K
1680
1680
1680
1680
1680
-
493K
528
528
528
528
528
-
Do the following:
Enter the data. Use the Eyring-Weibull model to calculate the parameters.
View all plots at once, including the Life vs. Stress plot, in ALTA's Side-by-Side Plots utility.
Solution
The file for this example is saved in the "Training Guide" folder and is named "Example7.ra6."
40
Create a new Data Folio. From the Data Type Expert select the appropriate data type and the appropriate
stress. Click OK to create the Folio. Enter the data into the Data Entry Spreadsheet and select Eyring as
the life-stress relationship and Weibull as the underlying life distribution.
http://ALTA.ReliaSoft.com
ALTA 6
Calculate the parameters by clicking the Calculate icon. The calculated parameters are shown next.
Plot the data. The Use Level Weibull Probability plot will appear, as shown below.
Eyr/Weib
Data 1
393
F=17 | S=23
90.00
Unreliability
50.00
10.00
5.00
User's Name
Company
6/11/02 15:32
1.00
10000.00
1.00E+5
Time
http://ALTA.ReliaSoft.com
41
ALTA 6
Obtain the Reliability vs. Time plot by selecting Reliability vs Time from the Special Plot Type dropdown menu. The Reliability vs. Time plot is shown next.
Reliability vs Time
1.00
Eyr/Weib
Data 1
393
F=17 | S=23
0.80
Reliability
0.60
0.40
0.20
User's Name
Company
6/3/02 16:17
0
0
4.00E+4
8.00E+4
Time
42
http://ALTA.ReliaSoft.com
1.20E+5
1.60E+5
2.00E+5
Next, select Pdf Plot from the Special Plot Type drop-down menu. The Probability Density Function
plot is shown next.
Eyr/Weib
Data 1
393
F=17 | S=23
8.00E-5
6.00E-5
f(t)
ALTA 6
4.00E-5
2.00E-5
User's Name
Company
6/3/02 16:17
0
0
6.00E+4
1.20E+5
1.80E+5
2.40E+5
3.00E+5
Time
http://ALTA.ReliaSoft.com
43
ALTA 6
Select Failure Rate vs Time from the Special Plot Type drop-down menu to obtain the Failure Rate vs.
Time plot. The Failure Rate vs. Time plot is shown next.
Eyr/Weib
Data 1
393
F=17 | S=23
8.00E-5
Failure Rate
6.00E-5
4.00E-5
2.00E-5
User's Name
Company
6/3/02 16:17
0
0
6.00E+4
1.20E+5
Time
44
http://ALTA.ReliaSoft.com
1.80E+5
2.40E+5
3.00E+5
ALTA 6
Select AF vs Stress from the Special Plot Type drop-down menu to obtain the Acceleration Factor vs.
Stress plot. The Acceleration Factor vs. Stress plot is shown next.
Eyr/Weib
Data 1
393
F=17 | S=23
Acceleration Factor
160.00
120.00
80.00
40.00
User's Name
Company
6/3/02 16:18
0
300.00
340.00
380.00
420.00
460.00
500.00
Temperature
If you would like to view all of these plots at the same time, along with the Life vs. Stress plot, you can use
the Side-by-Side Plots utility. Side-by-Side Plots is available from any Plot Sheet and uses the data, lifestress relationship and distribution from the linked Data Entry Spreadsheet. All plot properties, including
scaling and colors, is inherited from the current Plot Sheet. With Side-by-Side Plots, you can select from
three different life distributions, eight different life-stress relationships and nine different plot types to view
in the window.
or select Side-by-Side Plots from the Plot Options menu. The Side-by-Side Plots window will open.
On the Side-by-Side Plots Control Panel, click the Vary Plots button. Nine plot types will appear.
Select to view the Use Stress Level, Reliability, pdf, Failure Rate, Life vs. Stress and AF vs. Stress
plots by clicking the boxes to the left of the corresponding plot types.
http://ALTA.ReliaSoft.com
45
ALTA 6
Click the Refresh Plots button to refresh the plot area to reflect your selections. All six plots will appear
in the plot area, as shown next.
Close Side-by-Side Plots and return to the Folio by clicking the Close button. Save the Folio as
Example7.ra6 and close the Folio.
46
http://ALTA.ReliaSoft.com
ALTA 6
4.8 Example 8
In Example 6, the lower confidence level at a given percentile was estimated using the QCP. This example
demonstrates plotting confidence bounds. The use stress level for the following data set is 323K.
Number of Units
1
3
6
3
1
1
1
4
1
1
7
1
1
1
1
6
1
State
Failed
Failed
Suspended
Failed
Suspended
Suspended
Failed
Suspended
Suspended
Failed
Suspended
Suspended
Suspended
Suspended
Suspended
Suspended
Failed
Time-to-Failure
10512
11856
12191
2436
2436
2772
3109
4117
1175
1763
1882
1954
1129
1511
1607
1633
1895
Temperature
363
363
363
393
393
393
393
393
413
413
413
413
433
433
433
433
433
Do the following:
Determine the parameters for the data using the Arrhenius life-stress relationship and Lognormal
distribution.
Obtain the Use Level Lognormal Probability plot for the data with 90%, 2-sided Percentile (Type 1)
confidence bounds.
Solution
The file for this example is saved in the "Training Guide" folder and is named "Example8.ra6."
Create a new Data Folio. From the Data Type Expert, select the appropriate data sheet and stress type.
Set the stress properties and click OK to generate the Data Folio. Enter the data into the Data Entry
Spreadsheet.
http://ALTA.ReliaSoft.com
47
ALTA 6
Select Arrhenius as the life-stress relationship and Lognormal as the underlying life distribution.
Calculate the parameters. The next figure displays the results.
Create a Lognormal Probability plot by clicking the Plot icon. The Use Level Lognormal Probability
plot will appear on the Plot Sheet, as shown next.
Arrh/Log
Data 1
Unreliability
323
F=10 | S=30
50.00
10.00
5.00
User's Name
Company
6/3/02 16:22
1.00
10000.00
1.00E+5
Time
48
http://ALTA.ReliaSoft.com
ALTA 6
To plot the confidence bounds, select Confidence Bounds and then Show Confidence Bounds from
the Plot Options menu. The Confidence Bounds window will appear, as shown next.
Select Two Sided from the Sides options and Time (Type 1) from the Type options. Type 90 for the
Confidence Level (%) and click OK.
The Probability plot with confidence bounds is shown next.
Arrh/Log
Data 1
Unreliability
323
F=10 | S=30
CL: 90.00%
2 Sided-B
Time [T1]
50.00
10.00
5.00
User's Name
Company
6/3/02 16:27
1.00
10000.00
1.00E+5
Time
Notice that the legend now contains information about the plotted confidence bounds.
Save the Data Folio as Example8.ra6 and return to the Data Entry Spreadsheet by clicking the Data 1
page index tab. This Folio will be used for the next example.
http://ALTA.ReliaSoft.com
49
ALTA 6
4.9 Example 9
Using the data from Example 8, obtain tabulated values for the failure rate for 6 different mission end times.
The mission end times are 15,000 to 20,000 hr, incremented by 1,000 hr. This can easily be accomplished
using the Function Wizard within the General Spreadsheet. The General Spreadsheet is very similar to an
Excel spreadsheet. You can input formulas and edit the cells in a similar manner.
Solution
First, insert a General Spreadsheet into the Data Folio by selecting Insert General Spreadsheet from
the Folio menu. The Spreadsheet will look like the one shown below.
Once the Function Wizard is open, scroll down through the list of functions and select Failure Rate
Table as shown next. Note that the functions are in alphabetical order.
50
http://ALTA.ReliaSoft.com
ALTA 6
Input 15000 for the Start Time, 20000 for the End Time, 1000 for the Increment and 323 for the
Stress Level, as shown below.
Notice that at the bottom of the Function Wizard window the Data Source drop-down menu allows you
to select the Data Entry Spreadsheet you want to link the Function Wizard to. The Show Confidence
menu allows you to select to show confidence bounds. Keep these settings at the default settings of
Data 1 and None.
Click OK to obtain the table. The table will be placed into the General Spreadsheet, as shown next.
or select Save from the File menu. Leave the Data Folio open as it will be used in the next example.
http://ALTA.ReliaSoft.com
51
ALTA 6
4.10 Example 10
Using the data from Example 8, obtain the following plots:
Solution
The pdf vs. Stress and Reliability vs. Stress plots can be obtained by clicking the 3-D Plot icon,
or by selecting 3-D Plot from the Tools menu. When the 3-D Plot utility is activated, the following
window will appear first.
52
http://ALTA.ReliaSoft.com
ALTA 6
Select the pdf plot by clicking the corresponding box and clicking OK. The generated 3-D pdf plot is
shown next.
To rotate the plot, first click the plot. Then, while pressing the CTRL key, hold down the left mouse
button and move the mouse around to rotate the plot.
To create the Reliability vs. Stress plot, select Reliability from the Plot Type menu. The Reliability vs.
Stress plot is shown below and has been rotated to display a better view.
Close the 3-D Plot by clicking the Close icon located on the toolbar and close the Folio.
http://ALTA.ReliaSoft.com
53
ALTA 6
4.11 Example 11
The accelerating stress for an electronic component is temperature. To meet the specified reliability
requirement, the engineer was required to demonstrate with 90% confidence that 90% of the units will
continue to operate for 1000 hours under normal use conditions of 300K. To save time and money, the
engineer developed an accelerated life test designed to provide the desired reliability results in a shorter
period of time than would be possible with a test performed under normal stress conditions.
Consider the following times-to-failure and times-to-suspension of an electronic component that was placed
under an accelerated life test at three different stress levels: 353K, 373K and 393K. The normal use stress
level is 300K.
Stress Level
353 K
373 K
393 K
Observed Failures
245
110, 180, 200
50, 70, 88, 112, 140, 160
Do the following:
Determine the B(10) life of 1000 hours at a 90% lower 1-sided confidence for the use stress level
(300K).
Plot the Acceleration Factor vs. Stress plot for this example.
Solution
The file for this example is saved in the "Training Guide" folder and is named "Example11.ra6."
Create a new Data Entry Spreadsheet for time-to-failure and time-to-suspension data with one stress
column for temperature (in Kelvin, K). Enter the data into the Data Entry Spreadsheet.
Select Arrhenius as the life-stress relationship and Weibull as the underlying life distribution.
54
http://ALTA.ReliaSoft.com
ALTA 6
Notice that the Activation Energy (Ea) appears in the Results area. For this example, the Activation
Energy = 0.4653.
Next, activate the QCP. On the Confidence Bounds page, select Show Confidence and Show Lower
One-Sided. Make sure that the confidence level is set at 90%.
Return to the Basic Calculations page and select BX Information. Set the use stress level at 300 and the
BX% Information at 10. Calculate the B(10) life, as shown next.
The B(10) life demonstrated at the 90% lower 1-sided confidence is 1160, which is well above the
required 1000.
http://ALTA.ReliaSoft.com
55
ALTA 6
Plot the data. Select AF vs. Stress from the Special Plot Type drop-down menu.
Arrh/Weib
Data 1
300
F=10 | S=8
Acceleration Factor
72.00
54.00
36.00
18.00
User's Name
Company
6/11/02 11:19
0
300.00
320.00
340.00
360.00
Temperature
56
http://ALTA.ReliaSoft.com
380.00
400.00
ALTA 6
4.12 Example 12
Consider a chemical solution (e.g. ink formulation, medicine, etc.) that degrades with time. A quantitative
measure of the quality of the product can be obtained. This measure (QM) is said to be around 100 when the
product is first manufactured and decreases with product age. The minimum acceptable value for QM is 50.
Products with QM equal to or lower than 50 are considered to be "out of compliance" or failed.
Engineering analysis has indicated that at higher temperatures QM has a higher rate of decrease. Assuming
that the product's normal use temperature is 20 C (or 293K), the goal is to determine the shelf life of the
product via an accelerated degradation test. For the purpose of this analysis, "shelf life" is defined as the
time by which 10% of the products will have a QM that is out of compliance.
For this experiment, 15 samples of the product were tested, with 5 samples in each of three environments:
323K, 373K and 383K. Once a month, and for a period of seven months, the QM for each sample was
measured and recorded.
Do the following:
Solution
The file for this example is saved in the "Training Guide" folder and is named "Example12.ra6."
Activate the Degradation Analysis utility by clicking the Degradation Analysis icon,
When you first activate Degradation Analysis, the Open Degradation Analysis window will appear. For
this example, select Create a new Degradation base data set for this sheet and click OK. When you
create a new Degradation base data set, it will automatically attach itself to the current Data Entry
Spreadsheet so you can access it at any time. The attachment of this data set will be indicated by a red
tab that will appear in the Utility Attachment area of the Data Folio Control Panel's Main page. To open
the existing Degradation base data set file, you can either click the tab or right-click it to open a menu
and select Open Attachment.
You can also open the Degradation data set file that has already been created for this example. To open
the file, click the red Degradation tab in the Utility Attachment area of the Data Folio Control Panel's
Main page or select View existing Degradation data for this data sheet from the Open Degradation
Analysis window.
http://ALTA.ReliaSoft.com
57
ALTA 6
The Degradation Analysis window will open with a Data Entry Spreadsheet and a Control Panel. The
Control Panel, located on the right side of the window, allows you to select a Degradation Model,
Suspension Time, if any, and the Critical Degradation for the analysis.
In the Control Panel, select Exponential for the Degradation Model and type 50 for the Critical
Degradation. There is no Suspension Time for this example.
The Data sheet will look similar to the one shown next. Be sure the data set is entered as given in the
table. All of the data that are applicable for this example do not appear in the next figure, as there are
more than 21 rows of data.
Once you have entered the data in the Data sheet and made the specifications in the Control Panel, click
the Calculate icon on the toolbar,
58
http://ALTA.ReliaSoft.com
ALTA 6
Once you click Calculate, the utility will create two new sheets. The Output Parameters sheet, which
can be accessed by clicking its tab, shows the parameters for the degradation model for each of the data
points. For this example, the Output Parameters sheet contains the parameter values for each unit and
stress level, as shown next.
You can plot the data from the degradation analysis by clicking the Plot icon on the toolbar,
The results from the degradation analysis will be shown graphically on the Plot sheet. On the Control
Panel located on the right side of the Plot sheet, select to show all the units on the plot by clicking Select
All.
http://ALTA.ReliaSoft.com
59
ALTA 6
Your plot will look similar to the one shown in the following figure.
The Output Raw Data sheet, which returns the extrapolated time-to-failure points for the degradation
data you entered, was also created when you calculated the data set. Go to the Output Raw Data sheet
by clicking its tab.
Next, transfer the data from this sheet to the Data Folio by clicking the icon on the toolbar,
60
http://ALTA.ReliaSoft.com
ALTA 6
The data from Degradation Analysis will be transferred to the Data Folio. To analyze the data in the
Data Folio, select Arrhenius as the life-stress relationship and Weibull as the life distribution.
Calculate the parameters. The Data Folio will look like the one below.
Activate the Quick Calculation Pad. Perform a calculation for Warranty (Time) Information. Enter 293
for the use stress level and 90 for the required reliability. Click Calculate. The results are shown next.
Using the Degradation Analysis utility and the QCP, the shelf life is calculated to be about 15.6 months.
http://ALTA.ReliaSoft.com
61
ALTA 6
4.13 Example 13
Consider the data summarized in the following table.
Time-to-Failure
406 K
248
456
528
731
813
Temperature
416 K
164
176
289
319
340
543
426 K
92
105
155
184
219
235
This data illustrates a typical 3-stress level, constant stress accelerated stress. The stress in this example is
temperature and the Arrhenius life-stress relationship will be used. In this case, the stress profile is constant
and each stress level can be considered a different profile. The use stress level is 300.
Do the following:
Perform a likelihood ratio test to validate the assumption of a common shape parameter across three
stress levels.
Solution
The file for this example is saved in the "Training Guide" folder and is named "Example13.ra6."
Create a Data Entry Spreadsheet for non-grouped times-to-failure data with temperature as the stress.
Enter the data and calculate the parameters using the Arrhenius life-stress relationship and the Weibull
distribution.
62
http://ALTA.ReliaSoft.com
ALTA 6
Next, create a Probability plot of the data. Select to show the probability scales on the plot by selecting
Show Probability Scales from the Plot Options menu.
0.5
0.6
0.7
0.8
1.0
0.9
1.2
1.6
1.4
2.0
3.0
99.00
6.0
Probability Weibull
Arrh/Weib
Data 1
300
90.00
406
F=5 | S=0
416
F=6 | S=0
426
F=6 | S=0
Unreliability
50.00
10.00
5.00
User's Name
Company
6/4/02 09:23
1.00
10.00
100.00
1000.00
Time
You can also perform a likelihood ratio test to validate the assumption of a common shape parameter
across three stress levels. To do this, click its icon,
or select Likelihood Ratio Test from the Tools menu. The Likelihood Ratio Test utility will appear.
http://ALTA.ReliaSoft.com
63
ALTA 6
Enter .1 as the significance level and click Calculate. The results will appear, as shown next.
The value for the likelihood ratio test statistic, T, is less than the value of the chi-squared. Therefore, the
shape parameter estimates do not differ statistically at the 10% level and it can be re-stated that the
shape parameter remains constant across the three stress levels.
64
http://ALTA.ReliaSoft.com
The following examples use time-varying stresses, life-stress relationships and other utilities/
functionalities that are only available in ALTA 6 PRO. Thus, these examples can only be performed
using ALTA 6 PRO. Examples that can be performed using ALTA 6 Standard can be found in Chapter
4.
5.1 Example 14
An electronic component was subjected to a voltage stress, starting at 2V (use stress level) and
increased to 7V in stepwise increments. The following steps, in hours, were used to apply stress to the
products under test: 0 to 250, 2V; 250 to 350, 3V; 350 to 370, 4V; 370 to 380, 5V; 380 to 390, 6V; and
390 to 400, 7V.
Eleven units are available for the test. All eleven units are tested using the same stress profile. Units
that fail are removed from the test and their total time on test is recorded. The following times-tofailure are observed in the test, in hours: 280, 310, 330, 352, 360, 366, 371, 374, 378, 381 and 385.
Do the following:
1.
Determine the mean life (often called MTTF or MTBF) and B(10) life of these components at the
normal use stress level of 2V.
Solution
The file for this example is saved in the "Training Guide" folder and is named "Example14.ra6."
Create a new Data Folio for times-to-failure data with voltage as the stress type.
ALTA 6 PRO
Once the Data Folio has been created, activate the Stress Profile Explorer by clicking its icon,
or by selecting Stress Profiles from the Data menu. In the Stress Profile Explorer, you can define any
stress profile in segments. The stress for each of these segments can be a constant value (as in the case
of this example) or defined as a function of time.
In the Stress Profile Explorer, enter the data in the Spreadsheet area on the left side of the window, as
shown next.
66
http://ALTA.ReliaSoft.com
ALTA 6 PRO
Next, plot the stress profile in the Stress Profile Explorer by clicking the Refresh Plot icon,
or by selecting Refresh from the Plot menu. The plot will appear in the Plot area on the right side of the
window, as shown next.
http://ALTA.ReliaSoft.com
67
ALTA 6 PRO
Name the stress profile as Voltage Step by typing the name in the Profile Name input box at the top of
the window and clicking Add/Update Profile. The stress profile will appear in the Profiles area, as
shown next.
You can now attach the stress profile library to the current Data Folio by clicking the Set As Current
button,
or by selecting Set as Current from the File menu. A window will appear notifying you that the stress
profile library has been attached to the current Data Folio, as shown next.
Click OK. Notice that the library name changes from New to Attached in the profile library area.1 This
means that the stress profile will be available to assign to data points in the attached Data Folio.
Exit the Stress Profile Explorer by selecting Exit from the File menu and return to the current Data
Folio.
In the Data Folio, select Cumulative Damage as the life-stress relationship. On the Set Analysis page
of the control panel, select Weibull as the underlying distribution and Power as the cumulative damage
relation.
1.
Attached is the default name given to the stress profile library. You can change the name of the library when you
export the library to a file. To export the library, select Export Library from the File menu.
68
http://ALTA.ReliaSoft.com
ALTA 6 PRO
Enter the times-to-failure data in the Time to Failure column, as shown next.
Next, assign the Voltage Step stress profile to each data point by right-clicking a cell in the Voltage
column. A window will appear to display a list of all the stress profiles in the attached stress profile
library, as shown next.
Click to select the Voltage Step stress profile and click Add to assign the profile to the data point. The
stress profile name will appear in the Voltage cell. Repeat this step to assign the stress profile to the rest
of the data points.
http://ALTA.ReliaSoft.com
69
ALTA 6 PRO
Next, activate the QCP. Calculate the mean life for the voltage stress at the use stress level of 2.
70
http://ALTA.ReliaSoft.com
ALTA 6 PRO
http://ALTA.ReliaSoft.com
71
ALTA 6 PRO
5.2 Example 15
Consider a test in which multiple stresses were applied simultaneously to a particular automotive part in
order to precipitate failures more quickly than they would occur under normal use conditions. The engineers
responsible for the test were able to quantify the combination of applied stresses in terms of a "percentage
stress" as compared to typical stress levels (or assumed field conditions). In this scenario, the typical stress
(field or use stress) was defined as 100% and any combination of the test stresses was quantified as a
percentage over the typical stress. For example, if the combination of stresses on test was determined to be
two times higher than typical conditions, then the stress on test was said to be at 200%.
The test was set up and run as a step-stress test (i.e. the stresses were increased in a stepwise fashion) and the
time on test was measured in hours. The step-stress profile used was as follows: until 200 hours, the
equivalent applied stress was 125%; from 200 to 300 hrs, it was 175%; from 300 to 350 hrs, it was 200% and
from 350 to 375 hrs, it was 250%. The test was terminated after 375 hours and any units that were still
running after that point were right-censored (suspended). Additionally, and based on prior analysis/
knowledge, the engineers also stated that each hour on test under normal use conditions (i.e. at 100% stress
measure) was equivalent to approximately 100 miles of normal driving.
The test was conducted and the following times-to-failure and times-to-suspension under the stated stepstress profile were observed:
Times-to-Failure
252
280
320
328
335
354
361
362
368
72
http://ALTA.ReliaSoft.com
Times-to-Suspension
375
375
375
ALTA 6 PRO
After performing failure analysis on the failed parts, it was determined that the failure that occurred at 328
hrs was due to mechanisms other than the ones considered. That data point was therefore identified as a
suspension in the current analysis. The modified data set for this analysis was:
Times-to-Failure
252
280
320
335
354
361
362
368
Times-to-Suspension
328
375
375
375
Do the following:
1.
2.
Determine the B1 life for the part (i.e. time at which reliability is equal to 99%) at the typical operating
conditions (i.e. stress = 100%), in miles.
Solution
The file for this example is saved in the "Training Guide" folder and is named "Example15.ra6."
Create a new Data Folio for times-to-failure data with suspensions. Select Other as the stress type. In
the Stress Properties window, change the stress name to "Percent Stress" and set the use stress to 100.
Once the Data Folio has been created, activate the Stress Profile Explorer. The profile library for this
example is already attached to the Example14.ra6 file and will open when the Stress Profile Explorer is
activated.
In the Stress Profile Explorer, enter the data in the Spreadsheet area. Plot the stress profile by clicking
the Refresh Plot icon.
http://ALTA.ReliaSoft.com
73
ALTA 6 PRO
Name the profile as Automotive by typing the name in the Profile Name Input box and clicking Add/
Update Profile. The Stress Profile Explorer will look similar to the one shown next.
You can now attach the stress profile library to the current Data Folio by clicking the Set As Current
button. Exit the Stress Profile Explorer.
In the Data Folio, select Cumulative Damage as the life-stress relationship. On the Set Analysis page
of the Control Panel, select Weibull as the life distribution and Power as the cumulative damage
relation.
Enter the observed times and state (failed or suspended) into the Data Entry Spreadsheet. Next, assign
the Automotive stress profile to each data point by right-clicking each cell in the Percent Stress column
and selecting Automotive from the window that appears.
74
http://ALTA.ReliaSoft.com
ALTA 6 PRO
Next, generate an automated report of the step-stress analysis by clicking the ALTA Reports icon,
or by selecting ALTA Reports from the Tools menu. From the ALTA Reports window that appears,
select View or Print Standard ALTA Report and click OK.
The Auto Report Generator will open with a standard report based on the analysis in the Data Entry
Spreadsheet that includes the life-stress relationship, life distribution, use level, a Cox-Snell Residuals plot
and the raw data.
You can print a copy of this report by clicking the Print icon.
Close the Auto Report Generator and return to the Data Folio by clicking Close.
http://ALTA.ReliaSoft.com
75
ALTA 6 PRO
Activate the QCP to determine the B(1) life at the part's use stress level.
As shown above, the B(1) life is 657 hours. Based on the given multiplier, the B(1) life in miles would
then be 657 test - hr * 100 (miles/test - hr) = 65,700 miles.
76
http://ALTA.ReliaSoft.com
ALTA 6 PRO
5.3 Example 16
In many cases, the life of a product is a function of stress and some other engineering variable, such as
materials, vendors or operation type. For this type of product, ALTA 6 PRO provides the general log-linear
life-stress relationship, which allows you to analyze up to eight stress types and specify an underlying
relationship for each stress. This example demonstrates the use of this model.
A sample of electronic components were subjected to a quantitative accelerated life test in which three stress
types were applied to the units. The stress types included temperature, voltage and a third indicator variable
to describe whether the units were operated continuously or turned on and off. The stress profile for this test
is presented in the first table and the time-to-failure and time-to-suspension data are presented in the second
table. The normal use stress levels are 328K for temperature and 10V for voltage.
Profile
A
B
C
D
E
F
G
H
Temp (K)
358
358
378
378
378
378
398
398
Voltage (V)
12
12
12
12
16
16
12
12
Operation Type
On/Off
Continuous
On/Off
Continuous
On/Off
Continuous
On/Off
Continuous
Do the following:
Specify a transformation relationship for each stress type and estimate the parameters using the General
Log-Linear life-stress relationship.
Determine which of the stress types has a greater effect on the life of the product in terms of
accelerating failure.
Solution
The file for this example is saved in the "Training Guide" folder and is named "Example16.ra6."
Create a Data Entry Spreadsheet for grouped times-to-failure with suspensions data with three stress
columns for temperature, voltage and operation type.
http://ALTA.ReliaSoft.com
77
ALTA 6 PRO
Once the Data Folio has been created, insert a Subset ID column into the Data Entry Spreadsheet by
selecting Insert Subset ID Column and then Subset ID 1 from the Folio menu. Once the Subset ID
column has been inserted into the Data Folio, double-click the column heading to open the Change
Headings window. Rename the column heading to Stress Profile to identify the column as the one in
which the stress profile identifier will be entered.
Enter the data into the Data Entry Spreadsheet, as shown next. In the Operation Type column, enter 0 or
1 to indicate on/off or continuous operation, respectively.
Next, select Weibull as the underlying life distribution and General Log-Linear as the life-stress
relationship. When you select General Log-Linear from the Life-Stress Relationship drop-down box,
the Select Stress Column window will appear, as shown next.
78
http://ALTA.ReliaSoft.com
ALTA 6 PRO
Click the box next to each of the available stress columns so they will all be used in the calculations.
Click OK to close the window.
In the Stress Relation window, you can specify the transformation relationship for each stress type. For
this example, the temperature stress follows an exponential model, the voltage stress follows a power
model and no transformation will be performed on the operation type. The Stress Relation window
should look similar to the one shown next.
http://ALTA.ReliaSoft.com
79
ALTA 6 PRO
Calculate the parameters in the Data Entry Spreadsheet. The calculated parameters are shown next.
80
http://ALTA.ReliaSoft.com
ALTA 6 PRO
Next, plot the parameters on a Weibull-Probability plot. The plot will look similar to the one shown
next.
Probability Weibull
99.00
GLL/Weib
Data 1
328 / 10 / 0
358 / 12 / 0
F=2 | S=0
358 / 12 / 1
F=3 | S=20
378 / 12 / 0
F=4 | S=0
378 / 12 / 1
F=6 | S=14
378 / 16 / 0
F=4 | S=0
378 / 16 / 1
F=6 | S=10
398 / 12 / 0
F=4 | S=0
398 / 12 / 1
F=6 | S=7
90.00
Unreliability
50.00
10.00
5.00
User's Name
Company
8/2/01 14:41
1.00
10.00
100.00
1000.00
Time
The Weibull Probability plot can be used to examine the choice of an underlying life distribution and the
assumption of a common slope (shape parameter) at all stress levels. The linearity of the data and the fact
that the data for each stress level appears parallel reinforce the assumptions made.
http://ALTA.ReliaSoft.com
81
ALTA 6 PRO
Life vs Stress
10000.00
GLL/Weib
Data 1
Eta
328 / 10 / 0
F=35 | S=51
Life
1000.00
100.00
User's Name
Company
8/2/01 14:42
10.00
330.00
346.00
362.00
378.00
394.00
410.00
Temperature
Life vs. Stress plots can be very useful in assessing the effect of each stress on a product's failure. In this
case, since the life is a function of three stresses, three different Life vs. Stress plots are available. The
above plot was created by holding the voltage and operation type stresses constant at the desired use
level and varying the temperature stress. The method to create a Life vs. Stress plot where the voltage
varies is presented next.
Use the Use Stress Level window to create a Life vs. Stress plot for voltage by clicking the Use Stress
button on the Plot Sheet Control Panel.
In the Use Stress Level window, click the Vary option next to Voltage and click OK.
The Life vs. Stress plot for voltage will appear, as shown next.
82
http://ALTA.ReliaSoft.com
ALTA 6 PRO
Life vs Stress
1.00E+5
GLL/Weib
Data 1
Eta
328 / 10 / 0
F=35 | S=51
Life
10000.00
1000.00
User's Name
Company
8/2/01 14:44
100.00
10.00
100.00
Voltage
Finally, select to vary the operation type in the Use Stress Level window and click OK.
http://ALTA.ReliaSoft.com
83
ALTA 6 PRO
The Life vs. Stress plot for operation type will appear, as shown next.
Life vs Stress
10000.00
Life
GLL/Weib
Data 1
Eta
328 / 10 / 0
F=35 | S=51
User's Name
Company
6/11/02 14:14
1000.00
0
0.20
0.40
0.60
0.80
1.00
On/Off (0)
In this plot, the effects of two different operation types on life can be observed. It can be seen that the
on/off cycling has a greater effect on the life of the product in terms of accelerating failure than the
continuous operation. In other words, a higher reliability can be achieved by running the product
continuously.
84
http://ALTA.ReliaSoft.com
ALTA 6 PRO
5.4 Example 17
A specific consumer product (e.g. a mouthwash, shampoo, etc.) is made up of three main ingredients
(ingredients A, B and C) that have a characteristic (e.g. concentration) that may or may not change with
time. A quantitative measure of the characteristic can be obtained and this measure must be within a
specified range for compliance. If any measure is outside its specified range, then the product is out of
compliance and considered to be failed. There are no known dependencies among these ingredients. Thus,
they are assumed to be statistically independent.
For this study, 40 random specimens were stored at normal use conditions. At 3, 6, 9 and 12 months, 10
specimens were removed and measured. The measurement process is a destructive test (i.e. when the
specimen is opened for testing, the required readings are taken and that specimen is then disposed of).
Measurements for each ingredient at each time period are given in the following tables.
Low
High
Acceptable Range
A
B
142
155
156
185
C
110
135
If viewed from a "traditional reliability" perspective, the test in this example is not an accelerated test.
However, its analysis will require that we employ the fundamental principles of QALT analysis. The
measured value of each characteristic (as measured after each holding period) can be viewed as the random
variable (the time value in standard QALT analysis) affected by the aging process (the stress value). In other
words, the stress on each sample is the time in the holding cell and the random variable (what we
traditionally think of as time-to-failure) is the value of the measured characteristic.
http://ALTA.ReliaSoft.com
85
ALTA 6 PRO
Do the following:
The product has a shelf life of 24 months. Using the lognormal distribution along with a General LogLinear model, determine the probability that a given specimen will be out of compliance at/after this
time period.
Solution
The file for this example is saved in the "Training Guide" folder and is named "Example17.ra6."
Create a new Data Folio for times-to-failure data. Select Other as the stress type. Right-click the Other
button and in the Stress Properties window that appears, change the stress name to Age and set the use
stress to 1.
In the new Data Folio, rename the Time-to-Failure column to Char. Value. Enter the data for
ingredient A.
Select General Log-Linear as the life-stress relationship and open the Stress Relation window to
specify an untransformed relationship by selecting None.
Select the Lognormal distribution and calculate the parameters, as shown next. (Please note that the
figure below displays only 21 rows of data. Be sure to enter all of the data for this example as given in
the table on page 85. Your Data Entry Spreadsheet will contain 40 rows of data.)
Next, plot the Life vs. Stress plot with 90% 2-sided confidence intervals. Change the scaling of the plot
to reflect the acceptable range for product A.
86
http://ALTA.ReliaSoft.com
As can be seen from the plot shown next, no increase or decrease in the characteristic is noted. In other
words, age (at least up to the 12 months of observation) does not appear to affect the characteristic for
this ingredient.
Life vs Stress
156.00
GLL/Log
Data 1
Median
300
F=40 | S=0
CL: 90.00%
2 Sided-B
Time [T1]
Life
ALTA 6 PRO
User's Name
Company
6/4/02 10:14
142.00
0
6.00
12.00
18.00
24.00
30.00
Age
http://ALTA.ReliaSoft.com
87
ALTA 6 PRO
Next, use the QCP to determine the probability that ingredient A will be below the limit after 24
months, as shown next. Also determine the probability that ingredient A will be above the limit after 24
months.
The probability of ingredient A being below limit after 24 months is 0.02% and the probability of
ingredient A being above limit after 24 months is 0.03%
Insert a new Data Entry Spreadsheet into the Data Folio by selecting Insert Data Sheet from the Folio
menu.
In the new data sheet, enter the data for ingredient B. Calculate the parameters using the same settings
from the ingredient A analysis, as shown next. (Again, only 21 of the 40 rows of data are displayed in
this figure.)
88
http://ALTA.ReliaSoft.com
ALTA 6 PRO
Plot the Life vs. Stress plot with 90% 2-sided confidence intervals. Change the scaling of the plot to
reflect the acceptable range for product B. As can be seen from this plot, there is a noticeable decrease
in the characteristic.
Life vs Stress
185.00
Life
GLL/Log
Data 2
Median
300
F=40 | S=0
CL: 90.00%
2 Sided-B
Time [T1]
User's Name
Company
6/4/02 11:45
155.00
0
6.00
12.00
18.00
24.00
30.00
Age
Use the QCP to determine the probability that ingredient B will be below the limit, as shown next. Also
determine the probability of ingredient B being above limit after 24 months
http://ALTA.ReliaSoft.com
89
ALTA 6 PRO
It is found that the probability of ingredient B being below limit after 24 months is 0.23% and the
probability of ingredient B being above limit after 24 months is 0.00%
Enter the data for ingredient C and calculate the parameters using the same settings from the ingredient
A and B analyses, as shown next (with 21 of the 40 rows of data displayed).
90
http://ALTA.ReliaSoft.com
ALTA 6 PRO
Plot the Life vs. Stress plot with 90% 2-sided confidence intervals. Change the scaling of the plot to
reflect the acceptable range for product C. As can be seen from this plot, there is a noticeable increase in
the characteristic.
Life vs Stress
135.00
Life
GLL/Log
Data 3
Median
300
F=40 | S=0
CL: 90.00%
2 Sided-B
Time [T1]
User's Name
Company
6/4/02 11:52
110.00
0
4.00
8.00
12.00
16.00
20.00
Age
Use the QCP to determine the probability that ingredient C will be below the limit, as shown next. Also
use the QCP to determine the probability that ingredient C will be above the limit.
http://ALTA.ReliaSoft.com
91
ALTA 6 PRO
The probability of ingredient C being below limit after 24 months is 0.00% and the probability of
ingredient C being above limit after 24 months is 17.83%.
The results from the QCP calculations for each ingredient are shown next.
Ingredient A
Ingredient B
Ingredient C
Below Limit
0.02%
0.23%
0.00%
Above Limit
0.03%
0.00%
17.83%
In this case, the main contributing factor will be the the probability that ingredient C exceeds the limit,
which is significantly high. Having identified ingredient C as the main cause of failure, appropriate
corrective actions may be required.
92
http://ALTA.ReliaSoft.com
ALTA 6 PRO
5.5 Example 18
A new design is being considered for tanks that are intended to be used with heat exchangers for automotive
applications. It is desired to estimate whether the new design will meet its intended target before releasing it
to production. The target is a reliability of 90% at 2,000 hours of use, with a 90% confidence.
The stimulus that has the greatest effect on the life of the tanks is pressure, measured in pounds per square
inch (psi). In order to perform an accelerated life test on these tanks, groups of tanks were put on a pressure
cycling test. This test involves putting the tanks on a test fixture, pressurizing them to 5.5 psi and then
rapidly increasing the pressure to either 15 psi or 21 psi. The tanks are held at the elevated pressure for an
amount of time and then rapidly decreased to the original pressure. For each pressure cycle, the tank spends
80% of the time at the elevated pressure and 20% of the time at the default pressure of 5.5 psi. Since the
cycle repeats every 10 hours, each cycle can be thought of as a square wave. Thus, for each cycle, the tank is
held at 5.5 psi for 1 hour (one-tenth of the cycle), the pressure is rapidly increased to either 15 psi or 21 psi
and held for 8 hours (eight-tenths of the cycle), then the pressure is rapidly reduced to 5.5 psi and held for an
additional hour. The test fixture is capable of increasing and decreasing the pressure of the test units in a
rapid fashion, so much so that the transitions between the pressure levels can be considered simultaneous for
the purposes of this test.
The test was conducted with two groups of test units: 22 units were tested with the 15 psi profile and 10 units
were tested with the 21 psi profile. All of the units were tested to failure. That is, the stress profile was
repeated until all of the units on test had failed. The failure data set is contained in the following table:
Time-to-Failure (Hours)
for 15 psi Profile
2241
2493
3308
4018
4596
4845
5300
5691
5908
6466
6471
6720
6742
7718
8658
8927
9809
10155
10536
11194
11590
12774
Time-to-Failure (Hours)
for 21 psi Profile
1175
1369
4238
4392
4774
5109
5156
5422
7189
7551
-
http://ALTA.ReliaSoft.com
93
ALTA 6 PRO
As can be seen from the data, the range of failure times for the 15 psi pressure cycling test (2241 12774
hrs) is larger than the range for the 21 psi test (1175 7551 hrs), indicating an inverse relationship between
life and stress. However, the data set will need to be formally analyzed to see if this conjecture is correct.
The anticipated stress profile for the tank under normal use conditions for a mission of 2,000 hrs is 6 psi for
66.7% of its life, 11 psi for 25% of its life and 14 psi for 8.3% of its life.
Do the following:
Create two accelerated stress profiles for the 15 psi pressure cycling test and for the 21 psi pressure
cycling test.
Analyze the data with the Cumulative Damage-Weibull model and a Power relation to confirm whether
or not life is inversely proportional to stress.
Create a use stress profile to describe the anticipated use level stress for the tanks.
Estimate whether the new design will meet its intended target before releasing it to production.
Solution
The file for this example is saved in the "Training Guide" folder and is named "Example18.ra6."
Create a new Data Folio for times-to-failure data. Select Other as the stress type. Use the Stress
Properties window to change the stress name to Pressure and set the use stress level to 1.
Once the Data Folio has been created, activate the Stress Profile Explorer.
You can use the profile library that has already been created for this example or create a library according
to the steps described next. The profile library that has been created for this example is already attached to
the Example18.ra6 file and will open when the Stress Profile Explorer is activated.
94
http://ALTA.ReliaSoft.com
ALTA 6 PRO
In the Stress Profile Explorer, create a profile for the 15 psi pressure cycling test and plot the profile, as
shown next. Make sure to select Profile is Cyclical to specify that the test repeats itself until the units
fail.
Name the stress profile as 15 psi by typing the name in the Profile Name input box at the top of the
window. Click Add/Update Profile.
http://ALTA.ReliaSoft.com
95
ALTA 6 PRO
Create a new stress profile by selecting New Profile from the Profile menu. Enter the segments/stresses
for the 21 psi pressure cycling test and plot the profile. Remember to specify that the test repeats itself.
Name the stress profile 21 psi and add the profile, as shown next.
You can now attach the stress profile library that contains the profiles you created to the current Data
Folio by clicking the Set As Current button. Once the stress profiles have been attached to the current
Data Folio, exit the Stress Profile Explorer.
You will be prompted to export the new stress profile library as a separate file that will be accessible
from other Data Folios. For this example, de-select the Export Library option and click OK to proceed
without exporting the library.
In the Data Folio, select Cumulative Damage as the life-stress relationship. Set Weibull as the
underlying distribution and Power as the cumulative damage relation.
96
http://ALTA.ReliaSoft.com
ALTA 6 PRO
Enter the times-to-failure data from the failure data set and assign the corresponding stress profile (15
psi or 21 psi) to each data point, as shown next. (Please note that the figure below displays only 23 rows
of data. Be sure to enter all of the data for this example as given in the table on page 93. Your Data
Entry Spreadsheet will contain 32 rows of data.)
http://ALTA.ReliaSoft.com
97
ALTA 6 PRO
Next, create a Life vs. Stress plot, which can also be used to illustrate this hypothesis. (You may need to
change the scaling of the plot to 1 for the lower X-axis, 100 for the upper X-axis, 100 for the lower Yaxis and 100,000 for the upper Y-axis.) The negative slope of the lines in the graph indicates that life
decreases as pressure increases.
Life vs Stress
1.00E+5
CD/Weib
Data 1
Eta
Mean Life
300
F=32 | S=0
Life
10000.00
1000.00
User's Name
Company
6/5/02 12:17
100.00
1.00
10.00
100.00
Pressure
Now create the time-varying use stress profile for a mission of 2,000 hours. To do this, open the Stress
Profile Explorer and create a new profile within the existing stress profile library.
98
http://ALTA.ReliaSoft.com
ALTA 6 PRO
Enter the segments/stresses for the use stress profile, as shown next. You can see that this profile has the
highest stress in the middle of the profile, similar to the pressure cycles that were used for the test stress
profiles.
Name this profile as Center Cycle, as shown above, and add it to the stress profile library by clicking
Add/Update Profile.
Replace the static use stress value with the time-dependent stress profile that you have just created. To
do this, click the Use Stress button to open the Use Stress Level window and select Center Cycle from
the Use Stress Level drop-down box, as shown next.
Click OK to return to the Data Folio and then open the Plot sheet. Select the Reliability vs. Time plot
type.
http://ALTA.ReliaSoft.com
99
ALTA 6 PRO
The Reliability vs. Time plot for the analysis with the time-varying use stress profile is shown next.
1.00
CD/Weib
Data 1
Center Cycle
F=32 | S=0
0.99
Reliability
0.98
0.97
0.96
User's Name
Company
6/6/02 09:26
0.95
0
400.00
800.00
1200.00
Time
100
http://ALTA.ReliaSoft.com
1600.00
2000.00
ALTA 6 PRO
Use the QCP to calculate the reliability for the lower limit with 90% confidence bounds for a mission of
2000 hrs at the time-varying use stress profile, as shown next.
The 90% lower, one-sided reliability is calculated to be 91.02%. This exceeds the requirement of 90%
and the design can be released to production.
http://ALTA.ReliaSoft.com
101
ALTA 6 PRO
5.6 Example 19
This example requires you to have Weibull++ 6 installed on your computer.
In this example, we look at a way of using the proportional hazards model to help quantify an acceleration
factor that can link reliability results from in-house testing labs and the reliability experienced by items in
the field.
In order to determine the acceleration factor between the test and the field, test data and field data from the
same model of the product must be compared. The test and field data are presented in the following table.
Note that the time units differ between the in-house test and the field data; the test data set is in hours and the
field data set is in months. Rather than converting one set of data to the other's time units, this discrepancy
will be dealt with after the data have been analyzed.
Test Data
Times-to-Failure (in hours)
50
92
97
98
102
110
122
70
144
47
-
Field Data
Times-to-Failure (in months)
18
8
10
11
13
14
16
17
18
11 units suspended
at 19 months
Do the following:
Analyze the data with the Proportional Hazards-Weibull model to examine the relationship between lab
and field data.
Use the Probability and Contour plots to validate the assumptions regarding the nature of the
relationship between test lab and field use.
Determine the acceleration factor between the lab tests and the field.
Solution
The file for this example is saved in the "Training Guide" folder and is named "Example19.ra6."
Create a Data Entry Spreadsheet for times-to-failure data with suspensions and set Environment as the
stress factor.
Enter the test and field data in the Data Entry Spreadsheet, using 0 to denote field use and 1 to denote
lab testing.
102
http://ALTA.ReliaSoft.com
ALTA 6 PRO
Calculate the parameters using the Proportional Hazards model, as shown next.
Next, create a Weibull probability plot with 90% 2-sided confidence intervals to make sure that
assumptions regarding the nature of the relationship between test lab and field use are valid. Basically, if
the same failure modes are being attained at different stress levels, the slope of the Weibull probability
plot should be the same at the different stresses. This is shown in the following plot.
Probability Weibull
99.00
PH/Weib
Data 1
1
90.00
0
F=9 | S=11
1
F=10 | S=0
CL: 90.00%
2 Sided-B
Reliability [T2]
Unreliability
50.00
10.00
5.00
User's Name
Company
6/6/02 11:34
1.00
1.00
10.00
100.00
1000.00
Time
http://ALTA.ReliaSoft.com
103
ALTA 6 PRO
As the plot indicates, the data points for both environments conform closely to the calculated Beta value
of 3.53.
You can also use a Contour plot to examine the assumptions about the relationship. To do this, activate
Weibull++ by clicking its icon on the Control Panel,2
or by selecting Weibull++ 6 from the Tools menu. In the window that appears, select Transfer current
data from this sheet to a new Weibull++ 6 data sheet and click OK.
Once the data is transferred to Weibull++ 6, use the Batch Auto Run utility to put the field use data and
the test data in two separate Data Entry Spreadsheets. Open the Batch Auto Run by clicking its icon on
the Control Panel,
In the window that appears, click Select All Available. Both Subsets IDs (0 representing the field use
data and 1 representing the test data) will appear under the Selected Subset IDs area, as shown next.
Click OK.
2.
104
http://ALTA.ReliaSoft.com
ALTA 6 PRO
The field use data will appear in the Data Entry Spreadsheet labeled 01 with the parameters
calculated, as shown next.
The test data will appear in the Data Entry Spreadsheet labeled 11 with the parameters calculated, as
shown next.
Next, insert a Multiple Plot Sheet into the Weibull++ Data Folio by selecting Insert Multiple Plot
Sheet from the Folio menu.
Select Contour Plot as the plot type and select to plot both data sources, [W]01 and [W]11.
http://ALTA.ReliaSoft.com
105
ALTA 6 PRO
Refresh the plot. In the Select LK Contour Levels window that appears, select to plot the 1st, 2nd and
3rd levels at 95, 90 and 85 percent, as shown next.
Contour Plot
7.00
Weibull
01
F=9 / S=11
11
F=10 / S=0
5.60
: 95%
: 90%
: 85%
Beta
4.20
2.80
1.40
User's Name
Company
6/12/02 08:59
0
10.00
36.00
62.00
88.00
114.00
140.00
Eta
1=3.1702, 1=22.0028, =0.9900
2=3.1412, 2=104.1659, =0.9715
The Contour plot indicates that the values of Eta vary between the two environments, but the Beta
values fall within the same range. This further supports the assumption that the units are undergoing the
same failure modes in both the lab and the field.
106
http://ALTA.ReliaSoft.com
ALTA 6 PRO
In ALTA, open the QCP and calculate the acceleration factor between the lab and field data. The results
are shown next.
This calculation indicates that the acceleration factor from the field data (represented by stress level 1)
to the lab data (represented by stress level 0) is 4.7. In other words, the units in the field fail 4.7 times
faster than the units in the test lab.
Since the time units for the two data sets are different: the lab test times were in hours, while the field
data points were in months, we can say that 4.7 hours of time for this particular test is equivalent to one
month in the field. We have now quantified the acceleration factor between the lab testing and the field.
http://ALTA.ReliaSoft.com
107
ALTA 6 PRO
5.7 Example 20
This example requires you to have Weibull++ 6 installed on your computer.
Consider an electric motor that has been designed for a washing machine. The actual load sizes for the
motors in the field will vary depending on the way that the individual user employs each machine. The
manufacturer provides a warranty of 1,000 cycles for this motor. In order to make adequate preparations to
support the warranty, the manufacturer wants to estimate the percentage of returns that can be expected
during the warranty period. The life of the motor is clearly dependent on the applied load and the applied
load varies based on customer usage patterns. In other words, there is a distribution that governs customer
usage.
The manufacturer tested the life of the motor at different loads with the following results:
Cycles-to-Failure
Load (lb)
2386
6
3593
6
4045
6
6372
6
6448
6
1414
8
2147
8
3209
8
4026
8
4113
8
5117
8
6127
8
6352
8
819
12
1281
12
1441
12
1796
12
1856
12
2427
12
2645
12
2715
12
3671
12
4881
12
5 motors were suspended at 6500 cycles at the 6 lb load
1 motor was suspended at 6500 cycles at the 8 lb load
Customer usage information was also available from similar applications, obtained by conducting a survey
on a representative sample of customers and recording the sizes of the loads that they placed into their
washing machines. A Weibull distribution was fitted to the data and the following parameters were obtained:
Beta = 2
Eta = 6.2
108
http://ALTA.ReliaSoft.com
ALTA 6 PRO
Do the following:
Estimate the number of returns that can be expected during the warranty period.
Solution
The file for this example is saved in the Training Guide folder and is named Example20.ra6.
Create a Data Entry Spreadsheet for grouped times-to-failure data with suspensions with Load lb as
the stress type.
In the new Data Entry Spreadsheet, rename the Times-to-Failure column to Cycles-to-failure.
Enter the data into the data sheet and calculate the parameters using the Weibull distribution and Inverse
Power Law relationship, as shown next.
http://ALTA.ReliaSoft.com
109
ALTA 6 PRO
The next step is to determine the unreliability at 1,000 cycles using different loadings. Use the QCP to
calculate the unreliability for a load of 7 lbs., as shown next.
Continue to use the QCP to obtain values for loads of 5, 6, 8 and 9 lbs. These values are given (in
percent) in the next table:
Load (lb)
5
6
7
8
9
Using this data set, we can obtain the distribution of the percentage of units failing during the warranty
period of 1,000 cycles at each load size. This can be performed using the "Free Form" data type in
Weibull++ 6.
Keep ALTA open and activate Weibull++ 6.3 For this example, open the application outside of ALTA
(i.e. do not activate Weibull++ 6 from within ALTA ).
3.
110
http://ALTA.ReliaSoft.com
ALTA 6 PRO
In the new Weibull++ Data Entry Spreadsheet, enter the values that were obtained for each load and
calculate the parameters using the 2-parameter Weibull distribution with RRY for the parameter
estimation, as shown next.
Insert another Data Entry Spreadsheet into the same Weibull++ Data Folio (i.e. select Insert Data from
the Folio menu) for times-to-failure data.
In the new data sheet, set the 2-parameter Weibull distribution and click Calculate. In the window that
appears, set Beta = 2 and Eta = 6.2.
Now we have two distributions, one giving the percentage of units operating at each load size and the
second giving the percentage of units that fail at each load size during the warranty period of 1,000
cycles. The stress-strength interference analysis in Weibull++ can now be used to obtain the percentage
failing during warranty from the whole range of load sizes applied in the field.
To do this, activate the Stress Strength Interference Wizard by selecting Stress-Strength Calculations
from the Tools menu.
Set Data 1 as the Strength Set and Data 2 as the Stress Set. Click Calculate.
Based on this analysis, it is estimated that approximately 1.46% of the units will fail under warranty.
http://ALTA.ReliaSoft.com
111
6 Practice Questions
The following practice examples can be performed using ALTA 6 Standard and ALTA 6 PRO.1 The
answers to these practice questions can be found in Chapter 7.
Times-to-failure, hr
Suspension time, hr
1.
1.
2200 rpm
500
500
500
500
500
2500 rpm
290
310
316
339
451
-
2800 rpm
140
167
211
264
309
-
Determine how you would classify this data set, i.e. with or without suspensions, or in groups.
If you have a demonstration version of ALTA 6 or ALTA 6 PRO, you will not be able to enter your own data or
perform analyses for these examples. In addition, example files for these examples are not available. However,
example files are available for the examples in Chapter 4 and Chapter 5.
Practice Questions
ALTA 6
2.
3.
Determine the parameters for this data using the Inverse Power Law-Lognormal model.
4.
5.
6.
7.
Times-to-failure, hr
220 psi
165
177
238
290
320
340
341
380
449
544
230 psi
93
106
156
170
185
214
220
236
252
288
240 psi
72
73
99
124
134
150
182
186
190
228
250 psi
26
44
63
68
69
72
77
96
131
140
1.
Determine the parameters of the Inverse Power Law life-stress relationship and Weibull distribution.
2.
Obtain the Use Level Weibull Probability plot for this data set.
3.
Plot the 90%, 2-sided confidence bounds (Type II). Open RS Draw and save your plot as MyPlot1.wmf.
4.
5.
6.
7.
8.
9.
Obtain the Failure Rate vs. Time plot for these units. From the plot, what is the failure rate of these units
at 6000 hr?
114
http://ALTA.ReliaSoft.com
Practice Questions
ALTA 6
10. Obtain the Acceleration Factor vs. Stress plot. From the plot, what is the acceleration factor at 260 psi?
11. Save your file as QUEST2.ra6 and close the Data Folio.
Temperature,
K
378
378
378
378
378
378
378
378
398
398
398
398
Humidity
0.40
0.40
0.40
0.40
0.80
0.80
0.80
0.80
0.40
0.40
0.40
0.40
1.
Using the Temperature-Humidity life-stress relationship and Weibull distribution, determine the life for
a 90% reliability at the use stress levels.
2.
3.
4.
Change the entered humidity values from decimals to percentages (i.e. from 0.4 to 40). Repeat Step 1
(note that the use level humidity is now 20). Do you see any difference in your estimation?
Insert a new Data Entry Spreadsheet into this Data Folio for complete data with no censoring. (You
should now have an empty Data Entry Spreadsheet, as well as the one you saved from Question 2).
2.
Copy the data from Question 2 and paste it into the empty Data Entry Spreadsheet. Delete the last 10
entries for the 250 psi stress level. Use the Inverse Power Law life-stress relationship and Weibull distribution to calculate the parameters of the new data set.
3.
Insert another Data Entry Spreadsheet into this Data Folio for complete data with no censoring, and
again copy the data from Question 2 and paste it into the newly created Data Entry Spreadsheet. This
time, delete the last 20 entries for the 240 psi and 250 psi stress levels. Calculate the parameters for the
Inverse Power Law life-stress relationship and Weibull distribution.
4.
Plot the use level Weibull probability plot at 200 psi for each data set individually.
5.
Plot the 90% 2-sided confidence bounds and customize each plot to your liking.
6.
Insert a Multiple Plot Sheet into the Data Folio and select the Use Level Weibull Probability plot for all
three data sets to be displayed (double-click the data sets displayed in the Available Data Source(s)
box). What do you observe about the confidence bounds for each data set, especially Data 3 with the 20
failures?
http://ALTA.ReliaSoft.com
115
Practice Questions
ALTA 6
7.
Find the range of each confidence interval for each case using the QCP and for a 90% reliability at 200
psi. What happens to the confidence bounds when the number of test stresses increases? Why?
8.
Open QUEST1.ra6.
2.
Calculate the parameters using the Inverse Power Law life-stress relationship and Lognormal distribution.
3.
Plot the Failure Rate Function using ALTAs 3-D Plot utility.
4.
Press Ctrl and click the plot. Simultaneously move the mouse in the direction you would like to rotate
the plot.
5.
What is the behavior of the failure rate as time increases on a constant stress line?
6.
2.
3.
Activate ALTA Reports and select to launch the Report Work Center for custom reports. By default, the
Template page should open.
4.
On the Template page, create your own report by doing the following:
i. Insert a title and a few lines of text.
ii. Click Insert Field to activate the Function Wizard.
iii. Create a reliability table, starting at 1000 hr and ending at 2000 hr with an increment of 100 hr (the
use stress level is 1,500 rpm).
iv. Create a failure rate table, starting at 1000 hr and ending at 2000 hr with an increment of 100 hr (the
use stress level is 1,500 rpm).
v. Save you the Template page as MyTMPL.alt.
vi. Click the Editor button to compile the template and finalize your report.
vii. Include some graphics into the report. (At this point you should be in the Editor page of the Report
Work Center).
a.
Insert the plot you saved, MyPlot1.wmf, into the Editor page by selecting Picture then Insert
Picture from the Tools menu.
116
http://ALTA.ReliaSoft.com
Practice Questions
ALTA 6
2.
Generate a data set using the Eyring life-stress relationship and Lognormal distribution with std = 0.3, A
= 2, B = 5000. Create 25 data points for each of the 2 stress levels: 323 and 343.
3.
4.
Are the parameters equal to the ones used in the Monte Carlo generator? Comments?
5.
Generate another Monte Carlo data set using the same distribution and parameters.
http://ALTA.ReliaSoft.com
117
The purpose of this section is to provide answers to the Practice Questions in Chapter 6. Due to the
structure and process required for Questions 6 and 7, answers are not provided.
2.
Parameter values:
i. = 0.2485, K = 1.6896E-21, n = 5.3571
5.
6.
T = 1,008.6986 hours.
4.
7.
120
Increasing.
http://ALTA.ReliaSoft.com
ALTA 6