Remaining Useful Life prediction on C-MAPSS FD001
7.74RMSEDMHA-ATCN
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| DMHA-ATCNYear=2024, Pre-processing steps=Hybrid CNN2026.02 | 7.74 | — | — | — | — | — | — | |
| LSTM (with auto piece-wise)Year=2022, Pre-processing steps=Corr. analysis; median filter; norm; auto piece-wise RUL2026.02 | 7.78 | — | — | — | — | — | — | |
| CNN+LSTM+BiLSTMYear=2020, Pre-processing steps=Corr. analysis; min–max; RUL target2026.02 | 10.41 | — | — | — | — | — | — | |
| MsFormer2026.03 | 10.94 | — | — | 178.5 | — | — | — | |
| DFormerYear=2024, Pre-processing steps=Transformer variant2026.02 | 11.02 | — | — | — | — | — | — | |
| GA–RNN/LSTMYear=2021, Pre-processing steps=GA tuning; norm2026.02 | 11.19 | — | — | — | — | — | — | |
| CDSGYear=2023, Pre-processing steps=Graph structure; norm2026.02 | 11.26 | — | — | — | — | — | — | |
| MGLSNYear=2023, Pre-processing steps=Multi-granularity; norm2026.02 | 11.27 | — | — | — | — | — | — | |
| DVGTformer2026.03 | 11.33 | — | — | 179.75 | — | — | — | |
| MSAN2026.03 | 11.42 | — | — | 198 | — | — | — | |
| Multi-Scale CNNYear=2020, Pre-processing steps=Multi-scale conv; norm2026.02 | 11.44 | — | — | — | — | — | — | |
| MLEAN2026.03 | 11.48 | — | — | 186 | — | — | — | |
| HSMGNN2025.12 | 11.53 | 8.69 | 132.94 | — | — | — | — | |
| LSTM + multi-layer self-attnYear=2021, Pre-processing steps=Windows; norm; self-attn2026.02 | 11.56 | — | — | — | — | — | — | |
| FCSTGNN2025.12 | 11.62 | 9.16 | 135.02 | — | — | — | — | |
| GA+PredictorYear=2023, Pre-processing steps=GA tuning; norm2026.02 | 11.63 | — | — | — | — | — | — | |
| DiffRULYear=2024, Pre-processing steps=Diffusion-based RUL2026.02 | 11.71 | — | — | — | — | — | — | |
| Bi-level LSTMYear=2022, Pre-processing steps=Hierarchical LSTM2026.02 | 11.8 | — | — | — | — | — | — | |
| Attention-DCNNYear=2021, Pre-processing steps=Conv features; attention; norm2026.02 | 11.81 | — | — | — | — | — | — | |
| DTW-GPR2026.03 | 12.02 | — | — | 212.59 | — | — | — | |
| TF-SCN2026.03 | 12.05 | — | — | 219 | — | — | — | |
| BLS+TCNYear=2022, Pre-processing steps=Feature sel.; norm; piece-wise RUL2026.02 | 12.08 | — | — | — | — | — | — | |
| EAPNYear=2023, Pre-processing steps=Embedded attention; norm2026.02 | 12.11 | — | — | — | — | — | — | |
| EAPN2026.03 | 12.11 | — | — | 245.32 | — | — | — | |
| NSD-TGTN2026.03 | 12.13 | — | — | 226 | — | — | — | |
| IMDSSN2026.03 | 12.14 | — | — | 206.11 | — | — | — | |
| Multi-head CNN+LSTMYear=2020, Pre-processing steps=Feature sel.; RUL target2026.02 | 12.19 | — | — | — | — | — | — | |
| TATFA-TransformerYear=2024, Pre-processing steps=Transformer + attention2026.02 | 12.21 | — | — | — | — | — | — | |
| SAGDFN2025.12 | 12.22 | 7.96 | 149.26 | — | — | — | — | |
| Crossformer2025.12 | 12.32 | 8.4 | 151.96 | — | — | — | — | |
| AGCNNYear=2020, Pre-processing steps=Feature sel.; norm; RUL target2026.02 | 12.42 | — | — | — | — | — | — | |
| TS-MLLM2026.03 | 12.45 | — | — | 233.4 | — | — | — | |
| AMR-Net2026.03 | 12.49 | — | — | 269.08 | — | — | — | |
| BiGRU + Temporal AttnYear=2022, Pre-processing steps=Norm; temporal attn2026.02 | 12.56 | — | — | — | — | — | — | |
| One Fits AllBackbone=GPT-22026.03 | 12.61 | — | — | 310.98 | — | — | — | |
| MAGNN2025.12 | 12.63 | 12.71 | 159.52 | — | — | — | — | |
| Hybrid DL prognosticsYear=2021, Pre-processing steps=Fusion; norm2026.02 | 12.67 | — | — | — | — | — | — | |
| DeeBERT2026.03 | 12.75 | — | — | 243.64 | — | — | — | |
| FedLSTMYear=2022, Pre-processing steps=Federated learning; norm2026.02 | 13.33 | — | — | — | — | — | — | |
| Variational EncodingYear=2022, Pre-processing steps=Variational reg.; norm2026.02 | 13.42 | — | — | — | — | — | — | |
| CNN-SSEYear=2024, Pre-processing steps=CNN + squeeze-excite2026.02 | 13.46 | — | — | — | — | — | — | |
| One Fits AllBackbone=Qwen3-0.6B2026.03 | 13.62 | — | — | 259.4 | — | — | — | |
| BiLSTMYear=2018, Pre-processing steps=Feature sel.; norm; RUL target2026.02 | 13.65 | — | — | — | — | — | — | |
| Neural ODEYear=2023, Pre-processing steps=ODE layer; norm2026.02 | 13.65 | — | — | — | — | — | — | |
| BiLSTM2026.03 | 13.65 | — | — | 295 | — | — | — | |
| MSTSDNYear=2024, Pre-processing steps=Multi-scale two-stream; norm2026.02 | 13.67 | — | — | — | — | — | — | |
| KDnet2026.03 | 13.68 | — | — | 362.08 | — | — | — | |
| BiGRU2026.03 | 13.71 | — | — | 302.27 | — | — | — | |
| Bi-LSTM based Attention methodYear=2022, Pre-processing steps=RUL target function2026.02 | 13.78 | — | — | — | — | — | — | |
| C-Transformer2026.03 | 13.79 | — | — | 475.46 | — | — | — | |
| GAT-DAT2026.03 | 13.83 | — | — | 318.6 | — | — | — | |
| PE-Net2026.03 | 13.98 | — | — | 280.87 | — | — | — | |
| BiLSTM2026.03 | 14.12 | — | — | 296.02 | — | — | — | |
| MATT2025.12 | 14.5 | 11.25 | 210.25 | — | — | — | — | |
| Hybrid modelYear=2021, Pre-processing steps=Feature sel.; norm; piece-wise RUL2026.02 | 15.68 | — | — | — | — | — | — | |
| LSTMYear=2017, Pre-processing steps=Data normalization; RUL target2026.02 | 16.14 | — | — | — | — | — | — | |
| CNN+LSTMYear=2019, Pre-processing steps=Var. threshold; norm; HI2026.02 | 16.16 | — | — | — | — | — | — | |
| TFSCL2026.03 | 17.48 | — | — | 584.51 | — | — | — | |
| Bi_cLSTM (Ours, 4 blocks)Year=2025, Pre-processing steps=OS norm; z-score by mode; RUL cap=1252026.02 | 17.51 | — | — | — | — | — | — | |
| MegaCRN2025.12 | 17.72 | 12.7 | 314.04 | — | — | — | — | |
| CNNYear=2016, Pre-processing steps=Data normalization; RUL target2026.02 | 18.44 | — | — | — | — | — | — | |
| QAQL2026.06 | 20.86 | 14.18 | 435.28 | — | 12.16 | 68.19 | 5.28 | |
| BACE-RUL2025.03 | 21.81 | — | — | 72,099 | 15.25 | — | — | |
| Quantum VQL2026.06 | 23.49 | 22.47 | 551.88 | — | 17.19 | 92.38 | 7.81 | |
| Quantum DQN2026.06 | 24.13 | 30.29 | 582.29 | — | 21.54 | 145.18 | 8.21 | |
| Quantum AOA2026.06 | 26.18 | 37.48 | 685.59 | — | 20.11 | 175.41 | 7.42 | |
| RF2025.03 | 28.32 | — | — | 146,988 | 20.99 | — | — | |
| Quantum Eigensolver2026.06 | 28.53 | 36.15 | 814.27 | — | 30.12 | 218.74 | 10.96 | |
| SVM2025.03 | 29 | — | — | 177,192 | 22.42 | — | — | |
| Quantum SGD2026.06 | 29.07 | 39.19 | 845.24 | — | 24.01 | 276.27 | 9.85 | |
| DATE2025.03 | 29.77 | — | — | 135,990 | 34.58 | — | — | |
| Quantum Decision Tree2026.06 | 31.06 | 42.83 | 965.12 | — | 26.47 | 324.28 | 14.28 | |
| Quantum LSTM2026.06 | 31.21 | 40.86 | 914.29 | — | 31.23 | 282.38 | 12.2 | |
| LSTM2025.03 | 31.41 | — | — | 177,115 | 37.89 | — | — | |
| BRR2025.03 | 31.54 | — | — | 133,331 | 26.26 | — | — | |
| DCNN2025.03 | 32.76 | — | — | 95,329 | 27.71 | — | — | |
| Cox-PH2025.03 | 37.03 | — | — | 235,272 | 29.11 | — | — | |
| Weibull AFT2025.03 | 52.49 | — | — | 562,073 | 44.18 | — | — |