Reliability Prediction Model of Aircraft using Self-Organizing Map

The analysis of spare parts invalidation data that contains system failures is becoming increasingly important in the aircraft maintenance. This paper presents a systematic methodology to construct a reliability prediction model for aircraft reliability based on spare parts invalidation data. 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 the scatter spare parts invalidation data into a sequence model based on the time-to-failure data extracted from the repair registers. Its effectiveness is illustrated by Mean Time Between Failures (MTBF) study and analysis of real invalidation data. The proposed method can help proactively diagnose spare parts faults with a sufficient lead time before actual system failures to allow preventive maintenance to be scheduled thereby reducing the downtime costs.