Considering the gearbox is a primary source of noise and vibration in electrified powertrains, predicting key factors
such as gear face deviation and bearing misalignment is essential to reduce noise and vibration. This study aims to
predict gear face deviation and bearing misalignment using gear transmission errors through an artificial intelligence
model. The gear face deviation and bearing misalignment ranges were defined, and simulation data were generated for
model training. Consequently, we designed and trained a 1D-CNN-based artificial intelligence model using the collected
data. The model provided high prediction accuracy, with R² values exceeding 0.9 and RMSE values below 0.05 in most
cases. However, the prediction accuracy for the axial misalignment of Bearing 3 was relatively lower than that for the
other factors. In this study, we successfully developed a 1D-CNN model that predicts gear face deviations and bearing
misalignments from gear transmission errors with high accuracy. The proposed model is expected to contribute to
improved gearbox noise and vibration performance in electrified powertrains.