MTBF-Oriented Prediction Model for Airborne Equipment Reliability Based on SOM

The analysis of airborne equipment invalidation data that contains system failures is becoming increasingly important in the aircraft maintenance. However, carrying out an effective predictive maintenance plan, information about current airborne equipment reliability conditions must be understood to the decision-maker. In this paper, a systematic methodology to construct a prediction model for aircraft reliability based on airborne equipment invalidation data has been proposed. We take advantage of the large power of the Self-Organizing Map (SOM) technique developed by Teuvo Kohonen. SOM, that is an unsupervised neural network mapping a set of n-dimensional vectors to a two-dimensional topographic map, is used to combine their scatter data into a sequence model based on the time-to-failure data extracted from the repair registers. Its effectiveness is illustrated by the results of Mean Time Between Failures (MTBF) study and analysis. The method can help proactively diagnose airborne equipment faults with a sufficient lead time before actual system failures. It can allow preventive maintenance to be scheduled. Thereby it can reduce the downtime costs significantly.