Reliabilty Analysis - Quad Log - Engineering Specialized In Critical SystemsReliabilty Analysis The reliability analysis is a set of studies that characterizes the behavior of any given system, regarding the occurrences of faults. At this analysis, all universes of failures are considered, opposing to the safety analysis, where the focus is the study of the failures that tend to take the system to an unsafe or potentially unsafe state. The reliability prediction of a piece of equipment or of a fundamental part of a system is usually quantified by the calculation of the mean time between failures (MTBF). This calculation is usually based on standard MIL-HDBK 217-F-NOTICE 2, for electrical and electronic parts, and NPRD 95, for mechanical parts. Having the failures rate (λ), expressed in failures/h (or F/106h), it is possible to calculate the MTBF according to the following formula: MTBF = 1 / λ From the failures rate of fundamental blocks, models are developed considering the redundancies of the project analyzed and its properties regarding failures tolerance, being possible the evaluation of parameters such as mean times between maintenance (MTBA), mean times between failures on repairable systems (MTBF), mean times to failure on non-repairable systems (MTTF), probability of the success of a mission, etc. The reliability modeling may be evaluated by arranges of reliability blocks, Markov processes or by fault tree structures, getting to final rates which allow the evaluation of the adequacy of the analyzed system, regarding the desired performance rates. Availability Analysis An availability analysis is a study of the evaluation of the percentage of time in which a system works in a satisfactory way, without the occurrence of exaggerated degradation of operation. Therefore, the first step of this analysis is the definition of the implications premises for each kind of failure situation on the system operation, establishing effects groups, which vary from the insignificancy of the failure, where no operational trouble happens, to the massive failure, the one that makes the system inoperative. As an example, the objective regarding availability of an Operational Control Center of a subway chain has the order of 99,998%, what means, that the sum of all inactivity periods of an OCC during a year shall be less than 30 minutes. To evaluate if a real system reaches the availability goals, the fundamental rates of parts reliability, gathered during the reliability analysis, are used. It is also elaborated failure analyses focusing on the selection of those which make the system unavailable. Using mathematical modeling the failure grouping may be quantitatively evaluated, allowing the setting up of availability rates of the system as a whole and of its parts. Availability is a predominant factor in systems which may not have its operation stopped, either for social functions issues or because economic issues. Nowadays, when the demand for critical systems is strong, unavailability events may interfere, on a second moment, in the safety, due to popular dissatisfactions, followed by panic, depredations and social collapses with victims. Availability and Reliability Analysis
|