Some Contributions to the Methodology for Analyzing Failure Data.
GEORGE WASHINGTON UNIV WASHINGTON D C INST FOR MANAGEMENT SCIENCE AND ENGINEERING
Pagination or Media Count:
In this paper we deal with the important problem of failure data analysis. Our contributions can be classified into two broad categories. Under the first category comes the development of some goodness-of-fit tests for the Weibull distribution with unknown parameters. Our test statistics are functions of the empirical distribution function, and are based on an adaptation of some recent results on the theory of goodness-of-fit tests. Under the second category comes our proposal that the Lorenz curve methods of economic theory also be considered for use in the analysis of failure data. In the sequel, we point out several interesting connections between some well-known indicators in economic theory and a central concept in reliability theory. We feel that pointing out these connections between the two apparently different disciplines constitutes an important contribution of this paper. Author
- Statistics and Probability
- Manufacturing and Industrial Engineering and Control of Production Systems