ESCAFߝA New and Cheap System for Complex Reliability Analysis and Computation
- 1 October 1982
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Reliability
- Vol. R-31 (4) , 339-349
- https://doi.org/10.1109/TR.1982.5221371
Abstract
ESCAF stands for ``Ensemble de Simulation et de Calcul de Fiabilité'' - (``Electronic Simulator to Compute and Analyze Failures''). This paper describes the apparatus and illustrates, with examples, the multiple possibilities afforded to system design engineers, reliability engineers, and researchers: 1. Simulating and analyzing, directly or through fault trees, complex systems with as many as 416 components. 2. Simulating and analyzing networks with 2-way information flow. 3. Studying sequential circuits and flow networks. An exceptionally simple, cheap, and easy-to-use apparatus, ESCAF is employed typically to determine and list cut sets and (for s-coherent systems) minimal cut sets, or path sets and minimal path sets; to gauge the importance of each component or event by constructing histograms; to compute system failure probability or unavailability, for both s-coherent and non-coherent systems; and to detect potential weak-points in design. An optional diode plug programming matrix allows common-cause failures to be studied. To analyze a system with ESCAF, it is simulated very simply by standard integrated circuits mounted on one or more special cards. When inserted in the 440 × 226 × 470 mm basic ESCAF unit, these cards are automatically connected to a failure combination generator. This circuit generates all combinations of component failures (or fault tree basic events), first one-by-one, then two-by-two, and so on. The ESCAF processing unit analyzes the resulting system output states and derives the wanted information. The basic ESCAF unit is connected to one or two standard peripherals (visual display unit, printer, punched tape or diskette, etc.Keywords
This publication has 7 references indexed in Scilit:
- An Efficient Computational Technique for Evaluating the Cut/Tie Sets and Common-Cause Failures of Complex SystemsIEEE Transactions on Reliability, 1981
- Recursive Disjoint Products, Inclusion-Exclusion, and Min-Cut ApproximationsIEEE Transactions on Reliability, 1980
- An Algebraic Technique for Reliability EvaluationIEEE Transactions on Reliability, 1980
- Dagger-Sampling Monte Carlo For System Unavailability EvaluationIEEE Transactions on Reliability, 1980
- Fault Trees, Prime Implicants, and NoncoherenceIEEE Transactions on Reliability, 1980
- Difficulties in Fault-Tree Synthesis for Process PlantIEEE Transactions on Reliability, 1980
- Sequential Destruction Method for Monte Carlo Evaluation of System ReliabilityIEEE Transactions on Reliability, 1980