A methodology for assessing the reliability of ‘boxes’
| Autor: | Edward P. Budo, Stan Kaplan, Daniel Shaver, Daniel M. Cunha, James K. Liming, Andrew A. Dykes, Thomas C. Pritchett, Peter A. Mihalkanin |
|---|---|
| Rok vydání: | 1989 |
| Předmět: |
Operations research
Process (engineering) business.industry Industrial and Manufacturing Engineering Moment (mathematics) Bayes' theorem Conceptual framework Probability theory State (computer science) Artificial intelligence Safety Risk Reliability and Quality business Reliability (statistics) Mathematics |
| Zdroj: | Reliability Engineering & System Safety. 26:249-269 |
| ISSN: | 0951-8320 |
| DOI: | 10.1016/0951-8320(89)90014-8 |
| Popis: | In many practical situations, it is important to estimate the reliability of a system even though operational experience with the complete system is limited or nonexistent. In such situations, one is forced to fall back upon information available at the level of ‘boxes’ (i.e. of subunits of the complete system). Experience at this level often consists of results of various kinds of tests, all partially, but not totally, relevant to the performance of the box when it is part of the system. This paper lays out a language, conceptual framework, and methodology for dealing with such situations. The process is based on Bayes' theorem. The results are in the form of probability curves against the reliabilities of the individual boxes and the complete system. The probability curves show what state of knowledge or confidence we may logically have, at any moment, based upon the totality of information available at that time. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|