Accurate prediction of tool wear is critical for optimizing machining efficiency, minimizing unplanned downtime, and enabling predictive maintenance in intelligent manufacturing systems. However, developing computationally efficient models with high predictive accuracy remains a persistent challenge. To address this, we propose a novel pyramid-structured Long Short-Term Memory (LSTM) network for multi-step tool wear forecasting. The architecture features a hierarchical cascade of LSTM modules operating at progressively coarser temporal resolutions, enabling the model to capture both short-term fluctuations and long-term degradation trends. We further conduct a systematic ablation study to evaluate the impact of architectural depth and scale configuration on prediction performance. Experimental results demonstrate that the optimal model achieves a mean absolute percentage error below 0.5% on the validation set, a coefficient of determination close to 1, a mean absolute error of approximately 0.0005 mm, and a mean squared error lower than 5 × 10−6 mm2, indicating excellent predictive performance. These results underscore the model’s high precision and robustness. The proposed framework provides a reliable, data-driven solution for tool condition monitoring and supports optimal tool replacement scheduling in smart manufacturing environments.
ZhaoXZhengLShiM, et al. Controlling harmful milling vibration for robotic milling system based on the optimization of milling posture and spindle speed. Proc Inst Mech Eng Part B: J Eng Manuf2024; 239(12): 1747–1759.
2.
ZhangXZhaoZGuoZ, et al. Research on machining parameter optimization in finishing milling with multiple constraints. Proc Inst Mech Eng Part B: J Eng Manuf2022; 236: 968–980.
3.
WeiWCongRLiY, et al. Prediction of tool wear based on GA-BP neural network. Proc Inst Mech Eng Part B: J Eng Manuf2022; 236: 1564–1573.
4.
LiXLanMZhouY, et al. Parallel ResNet-BiGRU tool wear prediction model based on attention mechanism. Proc Inst Mech Eng Part B: J Eng Manuf2025; 0(0): 09544054251351358.
5.
PedrosoAFVSebbeNPVCostaRDFS, et al. INCONEL® alloy machining and tool wear finite element analysis assessment: an extended review. J Manuf Mater Process2024; 8: 37.
6.
FanGZhangJZhangG, et al. Finite element asnalysis of Ti1-xAlxN coated tools cutting performance and tool wear during Ti–6Al–4V milling. J Mater Res Technol2023; 25: 166–182.
7.
MethonGCourbonCAuzenatF, et al. Efficient predictive simulation of tool wear in face milling using an advanced 3D numerical approach. Wear2025; 570: 205946.
8.
HeJSunYGaoH, et al. On-line milling tool wear monitoring under practical machining conditions. Measurement2023; 222: 113621.
9.
HeZShiTXuanJ.Milling tool wear prediction using multi-sensor feature fusion based on stacked sparse autoencoders. Measurement2022; 190: 110719.
10.
TanselINArkanTTBaoWY, et al. Tool wear estimation in micro-machining: Part I: tool usage–cutting force relationship. Int J Mach Tools Manuf2000; 40: 599–608.
11.
ZhuLLiuC.Recent progress of chatter prediction, detection and suppression in milling. Mech Syst Signal Process2020; 143: 106840.
12.
SeemuangNMcLeayTSlatterT.Using spindle noise to monitor tool wear in a turning process. Int J Adv Manuf Technol2016; 86: 2781–2790.
13.
JangW-KKimD-WSeoY-H, et al. Tool-wear-estimation system in milling using multi-view CNN based on reflected infrared images. Sensors2023; 23(3): 1208.
14.
ZhangKZhouDZhouC, et al. Tool wear monitoring using a novel parallel BiLSTM model with multi-domain features for robotic milling Al7050-T7451 workpiece. Int J Adv Manuf Technol2023; 129: 1883–1899.
15.
WangJXiangZChengX, et al. Tool wear state identification based on SVM optimized by the improved northern goshawk optimization. Sensors2023; 23(20): 8591.
16.
ZegarraFCVargas-MachucaJCoronadoA.A comparative study of CNN, LSTM, BiLSTM, AND GRU architectures for tool wear prediction in milling processes. J Mach Eng2023; 23(4): 122–136.
17.
WarkeVKumarSBongaleA, et al. Robust tool wear prediction using multi-sensor fusion and time-domain features for the milling process using instance-based domain adaptation. Knowl Based Syst2024; 288: 111454.
18.
HeJXuYPanY, et al. Adaptive weighted generative adversarial network with attention mechanism: a transfer data augmentation method for tool wear prediction. Mech Syst Signal Process2024; 212: 111288.
19.
ZhengGLvXZhangX, et al. Wear monitoring of micro-milling tools based on Improved Siamese Neural Network. Proc Inst Mech Eng Part C: J Mech Eng Sci2024; 238(17): 8715–8729.
20.
WangCShenB.Auxiliary input-enhanced Siamese neural network: a robust tool wear prediction framework with improved feature extraction and generalization ability. Mech Syst Signal Process2024; 211: 111243.
21.
MaraniMZeinaliMSongmeneV, et al. Tool wear prediction in high-speed turning of a steel alloy using long short-term memory modelling. Measurement2021; 177: 109329.
22.
WuSLiYLiW, et al. Tool remaining useful life prediction considering wear state based on hybrid attention network. Proc Inst Mech Eng Part B: J Eng Manuf2024; 238(6–7): 837–850.
23.
QinBWangYLiuK, et al. A tool wear monitoring approach based on triplet long short-term memory neural networks. Proc Inst Mech Eng Part B: J Eng Manuf2024; 238(11): 1610–1619.
24.
WuXLiJJinY, et al. Modeling and analysis of tool wear prediction based on SVD and BiLSTM. Int J Adv Manuf Technol2020; 106: 4391–4399.
25.
ChengMJiaoLYanP, et al. Intelligent tool wear monitoring and multi-step prediction based on deep learning model. J Manuf Syst2022; 62: 286–300.
26.
DouJJiaoSXuC, et al. Unsupervised online prediction of tool wear values using force model coefficients in milling. Int J Adv Manuf Technol2020; 109: 1153–1166.
27.
WangRSongQPengY, et al. A milling tool wear monitoring method with sensing generalization capability. J Manuf Syst2023; 68: 25–41.
28.
ChengYGaiXJinY, et al. A new method based on a WOA-optimized support vector machine to predict the tool wear. Int J Adv Manuf Technol2022; 121: 6439–6452.
29.
ZhouYSunW.Tool wear condition monitoring in milling process based on current sensors. IEEE Access2020; 8: 95491–95502.
30.
ZhengGSunWZhangH, et al. Tool wear condition monitoring in milling process based on data fusion enhanced long short-term memory network under different cutting conditions. Maintenance Reliab2021; 23: 612–618.
31.
AnQTaoZXuX, et al. A data-driven model for milling tool remaining useful life prediction with convolutional and stacked LSTM network. Measurement2020; 154: 107461.
32.
HeJSunYYinC, et al. Cross-domain adaptation network based on attention mechanism for tool wear prediction. J Intell Manuf2023; 34: 3365–3387.
33.
LiKChenMLinY, et al. A novel adversarial domain adaptation transfer learning method for tool wear state prediction. Knowl Based Syst2022; 254: 109537.
34.
SunWZhouJSunB, et al. Markov transition field enhanced deep domain adaptation network for milling tool condition monitoring. Micromachines2022; 13(6): 873.
35.
LiYWangJHuangZ, et al. Physics-informed meta learning for machining tool wear prediction. J Manuf Syst2022; 62: 17–27.
36.
LiYLiuCHuaJ, et al. A novel method for accurately monitoring and predicting tool wear under varying cutting conditions based on meta-learning. CIRP Ann2019; 68: 487–490.
37.
HouYRagabMWangY, et al. Evidential domain adaptation for remaining useful life prediction with incomplete degradation. IEEE Trans Instrum Measur2025; 74: 1–11.
38.
GravesA (ed.). Long short-term memory. In: Supervised sequence labelling with recurrent neural networks. Springer Berlin Heidelberg, 2012, pp. 37–45.
39.
SakHSeniorABeaufaysF.Long short-term memory based recurrent neural network architectures for large vocabulary speech recognition. arXiv. arXiv:1402.1128, 2014.
40.
SunHZhangJMoR, et al. In-process tool condition forecasting based on a deep learning method. Robot Comput Integr Manuf2020; 64: 101924.
41.
VashishtRKPengQ.Engineering, Online chatter detection for milling operations using LSTM neural networks assisted by motor current signals of ball screw drives. J Manuf Sci Eng2021; 143: 011008.
42.
ZhangXHanCLuoM, et al. Tool wear monitoring for complex part milling based on deep learning. Appl Sci2020; 10: 6916.
43.
MareiMLiW.Cutting tool prognostics enabled by hybrid CNN-LSTM with transfer learning. Int J Adv Manuf Technol2022; 118: 817–836.
44.
LiuCLiYLiJ, et al. A meta-invariant feature space method for accurate tool wear prediction under cross conditions. IEEE Trans Indus Inform2021; 18: 922–931.
