Table 5 Impact analysis of different model components.

From: Comparison of deep learning models for predictive maintenance in industrial manufacturing systems using sensor data

Model variant

Accuracy (%)

F1-score (%)

CNN-LSTM

96.1 ± 0.2

95.2 ± 0.2

CNN-only

94.8 ± 0.3

93.6 ± 0.4

LSTM-only

93.5 ± 0.4

92.1 ± 0.5

FC-only

91.2 ± 0.6

89.4 ± 0.7

  1. Note: FC fully connected; results are averaged over five runs.
Back to article page