A review of FaultrEASE version 1.0
Many systems analysts will be surprised to encounter a program which is billed as a fault-tree development program "perform[ing] logical mathematical operations," but which does not perform Boolean reduction. The rather careful wording quoted in the introduction to this review can, in retr...
Gespeichert in:
Veröffentlicht in: | Risk analysis 1991-06, Vol.11 (2), p.343-8; discussion 349 |
---|---|
1. Verfasser: | |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | Many systems analysts will be surprised to encounter a program which is billed as a fault-tree development program "perform[ing] logical mathematical operations," but which does not perform Boolean reduction. The rather careful wording quoted in the introduction to this review can, in retrospect, be taken to refer to the fact that gates are quantified using formulas from the calculus of probabilities, and not to claim that Boolean reduction is performed. Since this program does not perform Boolean reduction, its use is limited to essentially graphical applications of the type illustrated in Fig. 1. For this limited application, the program has some features which make it attractive; it is easy to develop and print a passable drawing of a fault tree, and it is easy to do "what-if" analyses (looking at the effects of changing connections or statistics). However, for fault-tree analyses of even moderate complexity, a Boolean processor is necessary (a large fault tree for a real problem in which no events are repeated is arguably a pathological case). Many such algorithms exist on DOS machines, and most of them run within (and are limited to) the usual 640k memory limitation. To be fair, it has to be noted that some commercial algorithms of this type cost far, far more than FaultrEASE (their costs are measured in thousands of dollars rather than hundreds). |
---|---|
ISSN: | 0272-4332 |