The mostly working time of airborne electronic equipment is under preliminary depletion
failure phase, and inspection & maintenance at intervals can’t lower the failure probability. In this
paper, the law of airborne electronic equipment failure is introduced firstly. Then, methods for failure
prediction are summarized and analyzed. Finally, an example for predicting the airborne radar failure
using the Auto-Regressive (AR) and Support Vector Regression (SVR) model is presented. On this
basis, it is possible to achieve the goal that increases the reliability in working phase and establish a
more scientific maintenance system and to assure the safety of airborne electronic equipment.