Probability of failure per flight computations for aircraft systems

Aircraft systems are usually designed with highly reliable components by employing fault-tolerant architectures and redundancy management techniques. To further enhance the reliability and safety of aircraft, the components of these systems are checked or repaired at the end of periodic intervals of time, called maintenance check intervals. Depending on the criticality and functionality of the components, some are inspected prior to each flight, and others are inspected at different check intervals that coincide with airline scheduled maintenance. In addition, some components are monitored continuously using online automatic monitoring devices. Due to imperfect monitoring and maintenance check intervals that span multiple flights, failure of some redundant components cannot be identified immediately. Hence, a functional system can have latent failures in its redundant components. The probability of failure in such cases varies from flight to flight due to the different exposure times for the system components. This leads to an “average probability of failure per flight or flight hour” concept. This paper proposes a computationally efficient method for evaluating the average probability of failure per flight for aircraft systems considering generic system configurations, maintenance check intervals, backup units and latent failures, monitoring systems, and flight duration.