To solve the problem of incomplete single parameter evaluation in the performance evaluation of auxiliary power units (APU), this paper presented a method based on the logistic regression model for a complete performance evaluation of APU by integrating multiple parameters. Firstly, the source of APU parameter data was analyzed; a method for APU performance evaluation based on realtime message data was selected. Combined with the operating principle and real working condition of the APU, characteristic parameters that can reflect its performance were selected, and then the original characteristic parameters were reduced by principal component analysis (PCA). Secondly, the logistic regression model was built with the historical data of the APU in normal and severely degraded performances. Lastly, the health index (HI) of APU life cycle was obtained and verified based on the regression model. The results show that the APU performance health index curve can well reflect the performance changes of APU in each stage, and can provide technical reference for the subsequent development of condition based maintenance strategy. The results show that the health index curve of APU performance can well reflect its changes in different periods, and can provide technical references for the later formulation of oncondition maintenance strategies.