You are on page 1of 6

CHAPTER 70

Weibull and beta distributions: modeling machine life and duration of a project
Questions answered in this chapter:

How can I estimate the probability that a machine will work without failing for at least 20 hours? How can I estimate the probability that hanging drywall in a building will take more than 200 hours?

The Weibull random variable is a continuous random variable that is often used to model the lifetime of a machine. If you have data about how long similar machines have lasted in the past, you can estimate the two parameters (alpha and beta that dene a eibull random variable. ou can then use the WEIBULL.DIST function in Microsoft Excel 2013 to determine probabilities of interest such as an estimate of how long a machine will run without failing. The beta random variable is a continuous random variable thats often used to model the duration of an activity. Given estimates of the minimum duration, maximum duration, mean duration, and the standard deviation of the duration, you can use the BETA.DIST function in Excel 2013 to determine probabilities of interest.

Answers to this chapters questions


This section provides the answers to the questions that are listed at the beginning of the chapter. How can I estimate the probability that a machine will work without failing for at least 20 hours? Suppose you have observed the lifetime of seven similar machines. The data collected about the machines is contained in the eibullest.xlsx le, shown in igure 0-1.

691

FIGURE 70-1 Machine lifetime data.

Reliability engineers have found that the Weibull random variable is usually appropriate for modeling machine lifetimes. The Weibull random variable is specied by two parameters: alpha and beta. Based on the data, you can determine (using the AVERAGE and STDEV.S functions in cells B13 and B14) that, on average, a machine lasts 18.68 hours, with a standard deviation of 7.40 hours. By copying these values to cells G6 and G11 and running the Solver with the settings shown in Figure 70-2, you can nd estimates of alpha and beta that ensure that the Weibull random variable will have a mean and standard deviation matching the data. In this case, alpha equals 2.725 and beta equals 21.003, as you can see in Figure 70-3. Any value you enter for alpha and beta in cells E2 and E3 for a Weibull random variable yields a mean (computed in cell E6) and standard deviation (computed in cell E11). The Solver model varies alpha and beta until the mean and standard deviation of the Weibull distribution equal the mean and standard deviation of machine lifetime computed from your data. The Multistart option for the GRG Nonlinear engine is recommended for this. The Multistart engine works best with upper and lower bounds on the changing cells. A lower bound of 0.010 on alpha and beta will always work. An upper bound of 1,000 for alpha and beta was tried. If the Solver makes alpha or beta bump up against your upper bound, you should relax the bound.

692

CHAPTER 70

Weibull and beta distributions: modeling machine life and duration of a project

FIGURE 70-2 Solver settings for determining parameters for a Weibull random variable.

FIGURE 70-3 Estimates of alpha and beta for a Weibull random variable.

Weibull and beta distributions: modeling machine life and duration of a project CHAPTER 70

693

Heres the syntax of the WEIBULL.DIST function:


WEIBULL.DIST(x,alpha,beta,Cumulative)

When Cumulative equals True, this formula results in the probability that a Weibull random variable with parameters alpha and beta is less than or equal to x. Changing True to False yields the height of the Weibull probability density function (pdf). Remember from Chapter 66, An introduction to random variables, that the height of a pdf for any value x of a continuous random variable indicates the likelihood that the random variable assumes a value near x. Thus, if the Weibull density for 20 hours were twice the Weibull density for 10 hours, you would know that a machine is twice as likely to work for 20 hours before failing than to work for 10 hours before failing. Before answering some questions involving probabilities of interest, note that in prior versions of Excel, the WEIBULLDIST function yields the same results as WEIBULL.DIST:

What is the probability that a machine will last at least 20 hours? This probability (41.7 percent) is computed in cell G16 with the 1WEIBULL.DIST(20,alpha,beta,1) formula. Essentially, this formula computes the area under the Weibull pdf to the right of 20 hours by subtracting from 1 the area to the left of 20 hours. What is the probability that a machine will last from 15 through 30 hours? This probability (66.9 percent) is computed in cell G18 with the WEIBULL.DISTL(30,alpha,beta,TRUE) WEIBULL.DIST(15,alpha,beta,TRUE) formula. This formula nds the area under the Weibull pdf from 15 through 30 hours by computing the area to the left of 30 hours minus the area to the left of 15 hours. After you subtract the probability of a machine working without failure for less than 15 hours from the probability of a machine working without failure for less than or equal to 30 hours, you are left with the probability that a machine will work without failure from 15 through 30 hours.

How can I estimate the probability that hanging drywall in a building will take more than 200 hours? Since the development of the Polaris missile in the 1950s, project managers have modeled activity durations with the beta random variable. To specify a beta random variable, you need to specify a minimum value, a maximum value, and two parameters (alpha and beta). The data in the Beta.xlsx le (see Figure 70-4) can be used to estimate the parameters of a beta distribution. Suppose that you believe that the time needed to hang drywall in a building is sure to be between 0 and 600 hours. These are your minimum and maximum values, entered in cells C7 and C8. The F8:F22 cell range contains the lengths of time needed to hang drywall in 15 buildings of a similar size. In cell F5, the AVERAGE function was used to compute the mean time (78.49 hours) needed to hang drywall in these 15 buildings. In cell F6, the STDEV function was used to determine the standard deviation (47.97 hours) of the time needed to hang drywall in these buildings. Any choice of values of alpha and beta determine the shape of the beta distributions pdf and the mean and standard deviation for the corresponding beta random variable. If you can choose alpha and beta values to match the mean and standard deviation of the drywall installation times computed from this data, it seems reasonable that these alpha and beta values will yield probabilities consistent with the observed data. After you enter the mean and standard deviation for the drywall installation data in cells C9 and C10, 694
CHAPTER 70 Weibull and beta distributions: modeling machine life and duration of a project

the worksheet computes values for alpha (2.20) in cell C5 and beta (14.66) in cell C6 that ensure that the mean and standard deviation of the beta random variable match the mean and standard deviation of the data.

FIGURE 70-4 Determining probabilities with the beta random variable.

The BETA.DIST(x,alpha,beta,True_or_1,lower,upper) function determines the probability that a beta random variable ranging from lower through upper, with parameters alpha and beta, assumes a value less than or equal to x. The last two parameters are optional, and if they are omitted, Excel assumes lower equals 0 and upper equals 1. The BETA.DIST(x,alpha,beta,False_or_0,lower,upper) function returns the pdf for a random variable following a beta distribution. You can now use the BETA.DIST function to determine probabilities of interest. To compute the probability that hanging drywall will take at least 200 hours, you can use the formula in cell C15, 1BETA.DIST(200,alpha,beta,True_or_1,lower,upper). The result is 2.1 percent. This formula simply computes the probability that hanging drywall will take at least 200 hours as 1 probability dry walling takes less than or equal to 200 hours. The probability that hanging drywall will take at most 80 hours (58.3 percent) can be computed with the formula in cell C16, BETA.DIST(80,alpha,beta,True,lower,upper). And to compute the probability that the task will take from 30 through 150 hours (77.1 percent), in cell C17, use the BETA.DIS T(150,alpha,beta,True,lower,upper)BETA.DIST(30,alpha,beta,True,lower,upper) formula. This formula computes the probability that dry walling takes from 30 through 150 hours as the probability that dry walling takes less than or equal to 150 hours minus the probability that dry walling takes less than or equal to 30 hours. The difference between these probabilities counts only instances when dry walling takes from 30 through 150 hours.

Weibull and beta distributions: modeling machine life and duration of a project CHAPTER 70

695

Problems
The data for this chapters problems are contained in the Ch70data.xlsx le:
1.

In the Problem 1 worksheet, you are given data about the duration of a machines lifetime:


2.

What is the probability that the machine will last at least 10 hours? What is the probability that the machine will last from 1 through 5 hours? What is the probability that the machine will fail within 6 hours?

You need to clean your house today. In the Problem 2 worksheet, you are given data about how long it has taken to clean your house in the past. If you start cleaning at noon, what are the chances that youll be nished in time to leave at 7:00 P.M. for a movie?

696

CHAPTER 70

Weibull and beta distributions: modeling machine life and duration of a project

You might also like