[go: up one dir, main page]
Scribd
Upload
80 views1 page

Engineering Risk Analysis Task

This document provides instructions for an assignment with two questions analyzing time to failure data for diesel generators. Question 1 involves (1) preparing probability plots for normal, lognormal, gamma and Weibull distributions, (2) estimating distribution parameters using the probability paper and method of moments approaches, (3) determining the best fitting distribution, and (4) computing the mean lifetime using the best fit distribution and comparing it to the sample mean. Question 2 involves (1) estimating distribution parameters using maximum likelihood estimation, (2) computing means for each distribution, and (3) incorporating censored data to re-estimate the Weibull parameters and compare the resulting mean to the previous value from Question 2.

Uploaded by

Paula Rolim
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
80 views1 page

Engineering Risk Analysis Task

This document provides instructions for an assignment with two questions analyzing time to failure data for diesel generators. Question 1 involves (1) preparing probability plots for normal, lognormal, gamma and Weibull distributions, (2) estimating distribution parameters using the probability paper and method of moments approaches, (3) determining the best fitting distribution, and (4) computing the mean lifetime using the best fit distribution and comparing it to the sample mean. Question 2 involves (1) estimating distribution parameters using maximum likelihood estimation, (2) computing means for each distribution, and (3) incorporating censored data to re-estimate the Weibull parameters and compare the resulting mean to the previous value from Question 2.

Uploaded by

Paula Rolim
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

Assignment - 3

CivE 601: Engineering Risk Analysis – Fall 2019


Department of Civil and Environmental Engineering, University of Waterloo
Total Marks: 20 Due: October 4, 2019

Question 1 (10 marks)


The time to failure data for a sample of diesel generators is given in an Excel file, “Data_assignment_3”.
Prepare the probability papers of the data on normal, lognormal, gamma and the Weibull papers.
(1.1) Estimate the parameters of the fitted distributions using the PPP method.
(1.2) Compute the parameters of the four distributions (normal, lognormal, gamma and Weibull)
using the method of moments. Compare the results of the method of moment with those
obtained from the PPP method.
(1.3) Determine the best fit distribution for the time to failure data.
(1.4) Using the best fit distribution that you obtained in the previous step, compute the mean lifetime
of the diesel generators. Is this mean time to failure same as the mean of data set?

Question 2 (10 marks)


Use the time to failure data given in the Excel sheet in Question 1 to answer the following questions.
(2.1) Estimate the distribution parameters for normal, lognormal, gamma and Weibull distributions
using the maximum likelihood estimation (MLE) method.
(2.2) Compute the mean time to failure using the current estimates of the parameters of all four
distributions.
(2.3) A plant engineer later found additional information about censored lifetime data, (1) 10
generators of the operating age of 180 hours are still working in the plant, (2) 5 generators
failed between 400 and 500 hours of operation.
(2.3.1) Using the censored data, estimate the parameters of the Weibull lifetime distribution of
the diesel generator using MLE method.
(2.3.2) Compute the mean time to failure using the current estimates of the Weibull
parameters. Do you find any significant difference in the mean time to failure value as
compared to that you computed from the Weibull distribution in Question (2.2)?

You might also like