Tool Life Estimation and Optimization Using a Stacking Regressor and Whale Optimization Algorithm (WOA) in Micromilling

Abstract:

One of the primary drawbacks of the micromilling process is the limited lifespan of micro end mills, which significantly increases the cost of production. In addition to its brevity, the tool life is also stochastic in nature, making it challenging to predict and avoid untimely tool changes during the micromilling process. Despite the numerous attempts to estimate tool life, minimal attention has been paid to the feasibility of the ensemble machine learning models. Yet, they carry great potential for online tool status monitoring frameworks, which is the future of smart machining, a pillar of the Industry 4.0 industrial revolution. The aim of this work, therefore, was to develop an ensemble model for accurate tool life prediction and optimization. Data collected from a tool life experiment series involving the machining of a polymer-graphite composite were used to train and test a stacking regressor. The resultant model had an MAPE of 1.87%, an R² of 0.92 on both the training and testing data, as well as a R_adj² of 0.91, which are statistically significant. The model outperforms an empirical multilinear model, previously modelled using the same data, whose MAPE is 5.08 %, R² is 0.89 and R_adj² is 0.84. The stacking regressor is later used successfully in the whale optimization algorithm (WOA) to optimize tool life, and optimal cutting velocity (65.72 m/min), axial depth of cut (0.96 mm), and feed (26.16 µm) are proposed for the efficient micromilling of the polymer graphite composites, which are used for the fabrication of bipolar plates needed for hydrogen fuel cells.

• Periodical: Advances in Science and Technology (Volume 181)
• Pages: 169-179
• DOI: https://doi.org/10.4028/p-u1wRtZ
• Online since: August 2026
• Authors: Ogutu Isaya Elly*, Márton Takács
• Keywords: Machine Learning, Micromilling, Stacking Regressor, Tool Life, Whale Optimization Algorithm (WOA)

References

[1] B. Z. Balázs, N. Geier, M. Takács, and J. P. Davim, A review on micro-milling: recent advances and future trends, Int. J. Adv. Manuf. Technol., vol. 112, no. 3–4, p.655–684 (2021).

DOI: 10.1007/s00170-020-06445-w

[2] Ainur Munira Rosli, Ahmad Shahir Jamaludin, Mohd Nizar Muhd Razali, Recent Study on Hard to Machine Material Micromilling process, 九州大学学術情報リポジトリ Kyushu University Institutional Repository.

DOI: 10.5109/4480727

[3] A. Dadgari, D. Huo, and D. Swailes, Investigation on tool wear and tool life prediction in micro-milling of Ti-6Al-4V, Nanotech. Precis. Eng., vol. 1, no. 4, p.218–225 (2018).

DOI: 10.1016/j.npe.2018.12.005

[4] C. Christiand, G. Kiswanto, A. S. Baskoro, Z. Hasymi, and T. J. Ko, Tool Wear Monitoring in Micro-Milling Based on Digital Twin Technology with an Extended Kalman Filter, JMMP, vol. 8, no. 3, p.108 (2024).

DOI: 10.3390/jmmp8030108

[5] K. Vipindas, B. Kuriachen, and J. Mathew, Investigations into the effect of process parameters on surface roughness and burr formation during micro end milling of TI-6AL-4V, Int. J. Adv. Manuf. Technol., vol. 100, no. 5–8, p.1207–1222 (2019).

DOI: 10.1007/s00170-016-9210-3

[6] K. Vipindas, K. N. Anand, and J. Mathew, Effect of cutting edge radius on micro end milling: force analysis, surface roughness, and chip formation, Int. J. Adv. Manuf. Technol., vol. 97, no.1–4, p.711–722 (2018).

DOI: 10.1007/s00170-018-1877-1

[7] Y. Wang, B. Zou, J. Wang, Y. Wu, and C. Huang, Effect of the progressive tool wear on surface topography and chip formation in micro-milling of Ti–6Al–4V using Ti (C7N3)-based cermet micro-mill, Tribol. Int., vol. 141, p.105900 (2020).

DOI: 10.1016/j.triboint.2019.105900

[8] S. Saha, S. I. Ansary, S. Deb, and P. P. Bandyopadhyay, Influence of tool wear on chip-like burr formation during micro-milling, and image processing based measurement of inwardly-deflected burrs, Wear, vol. 530–531, p.205024 (2023).

DOI: 10.1016/j.wear.2023.205024

[9] T. Jithendra, S. S. Basha, A. Divya, and G. Rajyalakshmi, Machine learning technique ANFIS-COA for enhancing micro-milling performance by investigating the surface roughness and material removal rate, Int. J. Interact Des. Manuf. (2024).

DOI: 10.1007/s12008-024-02061-0

[10] C. Karagöz, B. Özlü, H. B. Ulaş, and H. Demir, RSM, ANN and ANFIS based analysis of machining conditions and cutting parameters in sustainable milling of AISI 316 Ti alloy, Multidiscip. Model. Mater. Struct. p.1–28 (2025).

DOI: 10.1108/MMMS-03-2025-0076

[11] J. Wang et al., Optimization for high-efficiency micromilling of titanium alloy based on an improved neural network and genetic algorithm, Mach. Sci. Technol., p.1–25 (2025).

DOI: 10.1080/10910344.2025.2551547

[12] A. Attanasio, E. Ceretti, and C. Giardini, SWARM Optimization of Force Model Parameters in Micromilling, Procedia CIRP, vol. 58, p.434–439 (2017).

DOI: 10.1016/j.procir.2017.03.248

[13] S. Mirjalili, Genetic Algorithm, in Evolutionary Algorithms and Neural Networks, vol. 780, in Studies in Computational Intelligence, vol. 780. , Cham: Springer International Publishing p.43–55 (2019).

DOI: 10.1007/978-3-319-93025-1_4

[14] U. L. Adizue, A. D. Tura, E. O. Isaya, B. Z. Farkas, and M. Takács, Surface quality prediction by machine learning methods and process parameter optimization in ultra-precision machining of AISI D2 using CBN tool, Int. J. Adv. Manuf. Technol., vol. 129, no. 3–4, p.1375–1394, (2023).

DOI: 10.1007/s00170-023-12366-1

[15] E. E.C. and van S. T.D., The micro milling of bi-poar plates -A tool life model, ISEM 2011 Proceedings.