Time Series Analysis and Forecasting of Failure Rate Processes.
Abstract
In this paper the author describes a new method for analyzing failure data. This method is particularly useful for analyzing data which is contaminated due to inspection, maintenance, or reporting procedures. The method is based on the notion that the estimated failure rate is the realization of a stochastic process which is referred to as the failure rate process. In order to be able to account for contaminations of the type discussed above, one allows the failure rate process to include periodic and trend components. The failure rate process is approximated by an appropriate time series model of the type suggested by BOX and JENKINS (1969). Such an approach is novel for reliability applications and has potential for a different direction of research in reliability theory.
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 16, 1974
- Accession Number
- ADA004203
Entities
People
- Nozer Singpurwalla
Organizations
- George Washington University