rerio Using Excl for Waibull Analysis = for Weibull Analysis by William W. Dorner Many people use Microsoft Excel on a daily basis. Yet few people realize the extent of Excel's analytical capabilities. Fewer still put these capabilities to work for process improvement, product improvement and profit Most Excel users are aware a Netscape users of the common formulas and charts, But with some creativity, users can produce ea to tools like control charts, Pareto charts and box-and-whisker plots (see "Using download this Excel for Data Analysis," Quality Digest, October 1997). And with a litle Zpped version guidance, users can employ more advanced statistical methods with Excel. This teat article presents a how-to approach for one such advanced technique-Weibull Welbullzip analysis. ‘You haven't tumed the page yet? Those of you who remain probably fall under one of two categories: those familiar with reliability data analysis, and Excel enthusiasts who are curious to learn one more way to exphoit this versatile software. I predict readers in both groups will be glad they stuck around. For the uninitiated, Weibull analysis is a method for modeling data sets containing values greater than zero, such as faihure data, Weibull analysis can make predictions about a product's life, compare the reliability of competing product designs, statistically establish warranty policies or proactively manage spare parts inventories, to name just a few common industrial applications. In academia, Weibull analysis has modeled such diverse phenomena as the length of labor strikes, AIDS mortality and earthquake probabilities. Learning by example Let's ignore the formulas for now and start by looking at an example of Weibull analysis in action, Imagine that you work for a toy company that wants to compare the reliability of two proposed designs for a jack- in-the-box spring housing. The desired reliability at 400,000 cycles is 0.90. In other words, the toy company would like 90 percent of the spring housings to survive at least 400,000 cycles. This reliability goal is expressed mathematically as R(400,000) = 0.90. Ten units were assembled with each of the two housing designs (Design A and Design B). These 20 units were tested until their spring housings failed. 1 Figure 1 shows the number of cycles before failure for each item tested. ww quality digest comangsininlsody_ woul et 9rerio The data in Figure 1 don't clearly indicate whether either design meets |[[IIENCIUTS 7 a the desired reliability goal, Both designs had at least one failure before |__Pesion Design 8 400,000 cycles, yet clearly the average number of cycles before failure [Sampte[ Cyctes [Sample] Cyctes exceeds 400,000 for both designs. A comparison of sample averages |-)—|-7s044 | 11 sza.naz using a Student's t test reveals no statistical difference between the 3 s0go77 [13 [650,570 average cycles for Design A and the average cycles for Design B (p- [2 —--zaer re Vaasa value = 0.965). But as a simple measure of central tendency, the sample 4 Bb ose 46 $s. om average gives no information about the spread or shape of the asks distribution of failure times. Could the two designs’ averages be the ops Te erat same, but their reliability be quite different? How can you be more scientific about comparing the reliability of the two proposed designs? Preparing to analyze ‘Modeling the data using Weibull analysis requires some preparation. For now, focus on the data ffom Design A. 1.Open Excel and into cell Al, type the label: Design A Cycles. Enter the failure data for Design A into cells A2:A11. Highlight cells A1:A11 and click on the Sort Ascending button to order the failure cycles from lowest to highest. 2.Jn cell BL, type the label: Rank. In cells B2:B11, type the integers 1-10 (see Figure 2) (ire A [6 c o. L f Design [Rank] Median Ranks | T/C TANGO Tn(Oesign 1 a cyces Median Rank) | Median rank | _ACyeles) 2| 384558] 1 | aoarso7a97 | 1o7ar6aoas | 2.c6seas08s | 12,8598499 3 | aa3.33i[ > [| o.r6346is3e [1.198402200 [-1.72326315 | 13.088457 4 sog.o77| + | o.2sgeissas [14506951 | 120207315 | 13.13838879 S| s1s.z0r[ 4 | 0.355769231 | 1352238806 | -0.82t666s1s [1315231239 BI 615.637] 5 | 0.451993077 | 182as61 ada | 0,508595394 | 1333007974 7 | 666,686] 6 | 0.sag0v6923 | 2717765957 | -0.730365445 | 1341007445, ‘8 726,044] 7 | 0.644730769 | 2.810810811 | 0,032024962 | 13.4953659, | 755,203] 8 | 0.7a0abasts | 3.8s18s18s2 | 0,299030932 | 13.53476835 To 07,863] 9 [-0.836538462 | 6.117647059 | 0593077217 | 1360714777 11 | 848,953] 10 | 0.932692308 [145714286 | 0.997688979 | 13.651 7591 3.In Colum C, put an estimate of the proportion of the population that will fail by the number of cycles listed in Column A. This can be accomplished using several different methods, the most common of which is median ranks. In cell C1, type the label Median Ranks. In cell C2, enter the formula: =((B2- 0.3)/(10+0.4)). Next, copy cell C2 down through cell C11. Note that in the formula for median ranks, the 10 in the denominator is the total number of Design A units tested. 4.Type into cell D1 the label: 1/(1-Median Rank). Then, in D2, enter the formula: D2 down through cell D11 1(1-C2). Copy cell $.Into cell El, enter the label: in(in(1/(1-Median Rank))). In cell E2, type the formula: =LN(LN(D2)). Copy cell E2 down through cell E11. 6.Finally, you'll need to transform the Cycles data, In cell FI, type the label: InDesign A Cycles). In cell F2, type the formula: =LN(A2). Copy cell F2 down through cell F11. 7.Again, compare your spreadsheet with Figure 2. After you confirm that everything is correct, save your workbook.rerio Using Excl for Waibull Analysis Fitting a line to the data Estimating Weibull Parameters At this point, you're ready to perform the Weibull analysis. The Why can we expect the graph of the thi t i In(Cycles) vs. the transformed median beauty of this method is that you can expect to see a straight ne MEY{es) we. Nhe Vanstormed when you plot the data in Column E vs. Column F.2 By performing a simple linear regression, you can obtain parameter With some effort, the Weibull cumulative estimates that will enable you to make inferences about Design distribution famtion can be transformed A's reliability.3 so that it appears in the familiar form of a straight line:Y=mX+b: Here's how: First, be sure that the Analysis TooIPak Add-In is loaded into Excel. From the menu bar, select Tools . Add-Ins. Click on the checkbox for Analysis ToolPak, and then click OK. To perform the simple linear regression: 1.While on the page you just created, fiom the menu bar, select Tools and Data Analysis. Scroll down and highlight "Regression and click OK. A data-entry window will pop up. 2.Under "Input Y Range," type: SE$1:SE$11 3.For "Input X Range," type: SFS1SFS11 4.Click to add a checkmark in the box for "Labels." whol ra] =Blax- fine 5.For "Output Options," select "New Worksheet Ply.” ‘Comparing this equation with the simple 6.Click to add a checkmark in the box for "Line Fit Plots." equation for a line, we see that the left ‘side of the equation corresponds to Y, Inx 7.Click OK. Excel will perform the regression and place the corresponds to X,@ corresponds to m, output on a new worksheet. and - In ecomresponds to b. Thus, when we perform the linear regression, the estimate for the Weibull parameter comes directly from the stope of the line. Reformatting the output The estimate for the® cz parameter must be calculated as follows: -|4| Before interpreting the output, you'll need to do some tidying up. ace (3 The columns do not automatically adjust to their optimal widths. To do this, within the worksheet that you just created, click on column heading A and drag to column heading I. Now double-click on the boundary to the right ofany column heading. Your table should look similar to Figure 3. ww quality digest comangsininlsody_ woul et aeSUMMARY OUTPUT Regression SeOTsES Multiple R KSquare Adjusted R Square Standard frror ‘Observations ANOVA Regression Residual Total Tnrerepe TagDesign ACTED Using Excel for Weibull Analyse O9RSIBITS O.97097813 O.967 35047 0.20147761 To z 7 5 3 Coaticent 371950557 42524822 Beta (or Shape Parameter) 3 Mis F TORGATSIOT_| TORGA9SI09 [PET GSATTD O.S2A7ASB17|_0.040893227 TL TEDGBEDT Significance F 1. 96274E-07 Tar 16.5083 708 TS.3601 467 Tower 95% es. 18218704 365508359) Prats T.82933007 T 3627607 Wlandard Error 346448033 239929377 Taper 9596 TAI. 2OS92AT EESTSBORTT [20_[ Alpha (or Characteristic Lite) ‘Now scroll to the right and click once on the graph, Stretch the graph by clicking and dragging on the handle in the lower right corner. From the menu bar, with the graph still selected, click View . Chart Window. Reformat the graph according to your preferences. It's best to use a solid line and no point markers for the Predicted line and delete the legend. Also, move the horizontal axis by clicking on the vertical axis with the right mouse button, selecting "Format Axis," clicking on the "Scale" tab and changing the "Value (X) Axis Crosses At" to -3 (see Figure 4). 25 132 194 InfDesign A Cycles) In cell A19, type the label: Beta (or Shape Parameter)~. In cell B19, type the formula: ~B18. In cell A20, type the label: Alpha (or Characteristic Life)=. In cell B20, type the formula: =EXP(-B17/B18). Your results should closely resemble Figure 3. For Design A, b=4.25 and a=693,380.4 ‘An identical analysis using the Design B data yields a@ g = 2.53 and ance =723,105. Interpreting the results The Weibull shape parameter, called, indicates whether the failure rate is increas decreasing, A g<1.0 indicates that the product has a decreasing failure rate. This scenario is typical of “infant mortality” and indicates that the product is filing during ts "burn-in" period. Ag =1.0 indicates a constant failure rate, Frequently, components that have survived bum-in will subsequently exhibit a ig, constant ortarsone Using Excl for Wet Anais constant failure rate. A g>1.0 indicates an increasing failure rate. This is typical of products that are ‘wearing out. Such is the case with the spring housings-both designs A and B have gz values much higher than 1.0. The housings fail due to fatigue, ie., they wear out, The Weibull characteristic lif, calleder, is a measure of the scale, or spread, in the distribution of data, It so happens that ce equals the number of cycles at which 63.2 percent of the product has failed. In other words, for a Weibull distribution R(c =0.368, regardless of the value of g. For example, with Design A housings, about 37 percent of the housings should survive at least 693,380 cycles. ‘While this is interesting, it stil doesn't reveal whether either jack-in-the-box design meets the reliability goal of R(400,000) = 0.90. For this, you need to know the formula for reliability assuming a Weibull distribution: 45 Rae where® x is the time (or number of cycles) until failure. The formula looks intimidating, but by simply plugging in the known values for et, g and x, you can obtain the desired reliability estimate o0,000742° @ 593380] _ (400,000 Design A= (400,001 723,105, 400,00) esien Bm? | =0.800 Computing the above formulas can be confusing and laborious using a calculator. Besides, you can't visualize or compare the reliability of each design for multiple cycle values. Excel provides a better way. Creating a reliability calculator worksheet 1 From your Design A regression output worksheet, highlight and copy cells A19:B20. Activate a new worksheet ply and locate the cursor in cell Al. Select Edit . Paste Special, click on Vahues, and click OK. This will paste your g and c labels and values into cells A1:B2 of the new worksheet. Resize the columns as needed. 2.In cell D1, type the label: Cycles. 3.ln cells D2:D11, type the values 100,000-1 milion in increments of 100,000. 4.In cell El, type the label: Survival Probability. 5.In cell E2, type the formula: WEIBULL(D2,SB$11,SB$2, TRUE). 6.Copy cell E2 down through cell El1. 7.1n cell F1, type the label: Reliability, 8.In cell F2, type the formula: =1-E2,rerio Using Excl for Waibull Analysis 9.Copy cell F2 down through cell FI1 10.Reformat cells as desired. Compare your worksheet with the top portion of Figure 5. a 6 T [Beta Cycles | Survival (or Shape Parameter) «| 4.25 Probability TAlpra (or Characteristic Life) =| 693,388] 1o0,000_| 0.0003 | 0.9997 200,900 | 10,0080 —| 0.9950 300,000—| 0.0280 | 0.9770 400,00 0.0919} 0-908 ‘300,000 0.220 | 0.7796 600,000 [0-416 | 0.3824 70.00 —| 0.6470 —| 0.3550 ‘300,000 [0.8407 [0.1593 900,000 0.9518 | 0.0487 000,000 [0.9913 [0.0087 Reliability] —Cyetes o.01 392,975 0 Ba3,674 os 636,127 7 0858 099 [735,056 You've now created a Weibull reizbilty calculator. You supply thocr, and cycles of interest, and Excel calculates the reliabilities for you. By merely changing the inputs in cells B1, B2 and D2:D11, you can get reliability estimates for any Weibull distribution of interest. Likewise, sometimes you'll need to compute the number of cycles (or time to failure) corresponding to a certain reliability level. For example, 99 percent of Design A housings will have failed by how many cycles? Unfortunately, Excel doesn’t have an inverse Weibull function. To perform this calculation (called solving for "critical values"), follow these steps: 1.On your Weibull reliability calculator worksheet, type in the label and values as shown in cells C13:C18 in Figure 5 2.In cell D13, type the label: Cycles. 3.In cell D14, enter the formula: =$BS2*(-LN(C14))*(1/$BS1). 4.Copy cell D14 down through D18. We find that for Design A of the jack-in-the-box, R(992,975)-0.01, or 99 percent of the housings will have failed by 992,975 cycles. ww quality digest comangsininlsody_ woul etrarore Using Ecol for Wobul Ana sis Creating a survival graph Perhaps the best way to compare the reliability of Design A with that of | 1] ices —[ Desian 8 [Desian 8 Design B is by using a survival graph. This line graph depicts the 3 [soo [p00 [sree survival probabilities of each housing type at various numbers of cycles. |4+ 1.00 | 0997 | 05714 Using the fornuis discussed above, enter the data into a new eT , ve Figure 6). Use the Chart Wizard to construct an X-Y [225000009870 —[-o.sar worksheet (see Figure 6). Use the Chart Wizard to construct an Xe [2{ Bow {oss fossa scatterplot, Select line styles of your choice and delete the point [ssn 0n0—| “ase —[- Ose markers, The resulting survival graph looks like Figure 7. in| 000 [aso [or . . . fa —son00—| “ure [oar Figure 7 allows a comprehensive comparison ofthe two designs’ fi —sston0 [888 [0.6068 survival rates. Note that at 400,000 eyoles, about 90 percent of Design [H4[ soon | ass2« 0360 A housings have survived, whereas only about 80 percent of Design B» [ig — sagan [oss 0-1 housings have survived. Therefore, for the stated reliability goal of Z| Zsuor [aze7s | os R(400,000) 20.90, Design A is clearly superior. However, about 10 fi] —wapor[oteaa tn percent of Design B housings will survive to 1 milion cycles, ws. fewer fof swoon {oon [oss than | percent of Design A. This graph clearly shows the importance of By] auams --00n2 FD defining the reliability goal in order to choose the more desirable design, [21[.0s0.o00 [0.002 | 0.0766 a[1. 00,000 | ~cco0e [0.0556 ps iso00—|-ony—[-n.aioe Awarranty example ae [1-200.000-|-no000—[ 00374 br [a.2s0,000 | -o.0000—[-o.o184 Having settled upon Design A as the superior altemative, suppose your 24{ 1auou0[o.couw —{-o.01 s “ en upen » SUPPOSE YOU 1>51350,000 | 0.0000 | 0.0078 company plans to offer a warranty on the jack-in-the-box, Of course, yf tawnano-ooom [ o.oos you would want to allocate suitable funds to honor the warranty, so as not to be blindsided by unexpected warranty costs, You've decided to set the warranty period so that no more than | percent of the units sold would fail before the warranty period expires. How can you determine what length of warranty to offer? (ene GaUabaagas The established Weibull model shows 99 percent of the housings should survive at least 235,056 cycles (see Figure 5). Market research shows that a heavily used jack-in-the-box is cycled 100 times per day. Wee find that 235,056 cycles equates to about 6.4 years of use. 235,056 cycles 4 years® cyeles 5 days 00 2 at ‘Armed with this information, and knowing that the competition only offers a two-year warranty on its, ww quality digest comangsininlsody_ woul ettarsone Using Excl for Wet Anais jack-in-the-boxes, your company might choose to be conservative and offer a five-year warranty. This ‘would ensure domination of the competition fiom a marketing standpoint, yet stil alow for warranty costs to stay at or below the desired le The above example is somewhat simplistic, Interested readers can find more sophisticated illustrations of warranty strategy using Weibull analysis in academic articles, such as Jayprakash Patankar and Amitava Mitra's "Effects of Warranty Execution on Warranty Reserve Costs" (Management Science, 1995). Abrief statistics overview Weibull analysis involves fitting a data set to the following cumulative distribution fimetion (cdi):5 a ° F(x)=1-¢ lel vforx>0 Confusion has arisen in the past due to the lack of standardized nomenclature for the Weibull cdf. Its creator, Waloddi Weibull, himself published multiple versions of this formula using different nomenclatures. ‘Arthur Hallinan Jr. provides an excellent history of the various forms of the Weibull distribution in "A. Review of the Weibull Distribution" (Journal of Quality Technology, 1993). The format above is the most commonly accepted one. Unfortunately, in Excel, the "Help" screen for the "=WEIBULL function gives the formula with the ce and i parameters reversed (ie., the characteristic life is labeled g and the shape parameter is labeleda) Conclusion The Weibull distribution's strength is its versatility. Depending on the parameters’ values, the Weibull distribution can approximate an exponential, a normal or a skewed distribution, The Weibull distribution's virtualy limitless versatility is matched by Excel's countless capabilities. An astute data analyst who understands the theory behind a given analysis can often get results from Excel that others might assume require specialized statistical software. With Excel, Weibull analysis les well within reach for most engineers with a statistics background. For more information The Excel fle used in this article and an explanation of estimating Weibull parameters are available from our Web site at www.qualitydigest.com/jan99/htmleibull html Notes 1. For simplicity, this article deals with complete failure data, ie., all samples were tested until they failed. Inpractice, reliability data analysis frequently involves censored data, or samples for which, for one reason or another, failure times are unknown. Often, tests are suspended before all samples fail, Or perhaps items ‘may fil due to a cause other than the one being studied. The issues involved in analyzing and interpreting censored data are complex. Improper analysis of censored data can yield misleading results, which Margaret Mackisack and Ronald Stillman point out in "A Cautionary Tale About Weibull Analysis" (/EEE Transactions on Reliability, 1996). For further technical details about analyzing censored life data, readers also can consult Wayne Nelson's book Applied Life Data Analysis (John Wiley & Sons, 1982) or William Meeker and Luis Escobar's book ww quality digest comangsininlsody_ woul et a9rerio Using Excl for Waibull Analysis Statistical Methods for Reliability Data (John Wiley & Sons, 1998). 2. Fora fill explanation of why you can expect a straight line, ww. vdiges n99/h ib ¢ this article at our Web site: 3. Many methods exist for estimating Weibull distribution parameters from a set of data. This article uses the method called probability plotting. Readers interested in other methods, such as maximum likelihood estimation or hazard plotting, should consult Nelson's book, Meeker and Escobar's book or Bryan Dodson's book Weibull Analysis with Software (ASQ Quality Press, 1994). 4, Some software packages may give slightly different parameter estimates than the ones in this article That is because these applications regress the transformed median ranks (¥) on the transformed lifetimes (X) rather than vice versa, "Simulation studies show that Yon X regression produces almost double the bias in the estimation of the shape parameter as the X on ¥ regression," according to Dodson. Moreover, the universal convention for displaying a Weibull probability plot is to depict "In(ifetime)" on the horizontal axis, The regression method presented in this article automatically generates the plot in this standard format. 5. The type of Weibull distribution discussed in this article is called the two-parameter Weibull distribution. This simple form is adequate for a majority of Weibull analysis scenarios. However, if the transformed failure data plot has a curved rather than a straight line appearance, or if g is found to be greater than 6.0, then a third parameter may be needed to adequately model the data. The third parameter, included in the aptly named three-parameter Weibull distribution, effectively shifts the entire distribution to the right. This Jocation parameter is most commonly called. (the Greek letter gamma). In practice, "can be interpreted as the earliest possible time at which failure may occur. Of course, #* may never be larger than the value of the earliest failure from the data set. Readers who encounter a curved regression plot or af value greater than 6.0 should consult Hallinan's article or John McCoo!’ article "Inference on the Weibull Location Parameter" (Journal of Quality Technology, 1998) for guidance on fitting a three-parameter Weibull model. About the author William W. Dorner is a senior quality/process engineer at Best Access Systems in Indianapolis, He is 4 Certified Quality Engineer and a member of ASQ and ASA. For more information, you may e-mail Domer at wdorner’@qualitydigest.com [QD Online] [Choose CMM] [CMM Guide] [CE Mark] [weibull] Copytight 1988 QC! Intemational. All rights served. Quality Digest can be reached by phone at (530) 893-4095, E-mail: lick Here ww quality digest comangsininlsody_ woul et sre