International Journal of Engineering and Manufacturing (IJEM)
IJEM Vol. 15, No. 6, 8 Dec. 2025
Cover page and Table of Contents: PDF (size: 1147KB)
Benchmarking Taguchi and Deep Neural Network Approaches for Fiber-Laser Micromachining of Stainless Steel: Multi-Objective Optimization of Kerf, HAZ, and Edge Integrity
PDF (1147KB), PP.16-32
Views: 0 Downloads: 0
Author(s)
Index Terms
Laser Micromachining, Stainless Steel Cutting, Process Parameter Optimization, Taguchi Method, ANN, CQI, SNR, Comparative Modelling, Intelligent Manufacturing, Predictive Modelling in Laser Processing
Abstract
Laser micromachining has become an essential tool in precision manufacturing due to its non-contact nature, high spatial resolution, and capability to produce intricate micro-features. However, identifying the optimal combination of process parameters remains challenging because of the nonlinear and interdependent effects of laser power, scanning speed, and pulse frequency on cut quality. In this study, a comparative framework is presented that benchmarks the Taguchi Design of Experiments (DoE) against a Deep Neural Network (DNN) model to predict and optimize the micromachining performance of stainless steel. A unified Cut Quality Index (CQI) was developed by combining three critical responses kerf width, heat-affected zone (HAZ), and edge chipping into a single measure of overall cut integrity. A physics-consistent dataset of 75 samples, comprising 20 literature-based and 55 synthetically generated data points, was constructed to ensure both experimental realism and statistical diversity. The Taguchi analysis using an L18 orthogonal array identified the optimal parameters as 80 W laser power, 250 mm/s scanning speed, and 60 kHz pulse frequency, corresponding to the highest signal-to-noise ratio and thermally balanced operation. The DNN model achieved strong predictive accuracy (R² ≈ 0.92–0.94), effectively capturing nonlinear parameter interactions without overfitting. The results demonstrate that while the Taguchi method efficiently identifies robust process windows with minimal experimentation, the DNN extends predictive capability across continuous, untested regions of the process space. Collectively, these findings establish a physics-informed, data-driven comparative framework for intelligent optimization of laser micromachining, with direct relevance to aerospace, biomedical, and precision micro-engineering applications.
Cite This Paper
Aswin Karkadakatil, "Benchmarking Taguchi and Deep Neural Network Approaches for Fiber-Laser Micromachining of Stainless Steel: Multi-Objective Optimization of Kerf, HAZ, and Edge Integrity", International Journal of Engineering and Manufacturing (IJEM), Vol.15, No.6, pp. 16-32, 2025. DOI:10.5815/ijem.2025.06.02
Reference
[1]Thejasree, P., Natarajan, M., Khan, M.A. et al. Application of a hybrid Taguchi grey approach for determining the optimal parameters on laser beam welding of dissimilar metals. Int. J. Interact. Des. Manuf., 19, 175–184 (2025). https://doi.org/10.1007/s12008-023-01588-y
[2]Mahrous, A.B. ED., Elsad, R.A. & El-Mahalawy, M. Multi-optimization of laser drilling of GFRP composites via TOPSIS approach. Discov. Appl. Sci., 7, 786 (2025). https://doi.org/10.1007/s42452-025-07281-x
[3]Tamrin, K., Nukman, Y., Choudhury, I., & Shirley, S. (2015). Multiple-objective optimization in precision laser cutting of different thermoplastics. Opt. Lasers Eng., 67, 57–65. https://doi.org/10.1016/j.optlaseng.2014.11.001
[4]Eltawahni, H., Benyounis, K., Olabi, A.G. (2015). High power CO₂ laser cutting for advanced materials – review. Ref. Module in Materials Sci & Eng. https://doi.org/10.1016/B978-0-12-803581-8.04019-4
[5]Aydın, K., & Uğur, L. (2025). Prediction of kerf and groove widths in CO₂ laser cutting process of PMMA using experimental and machine learning methods. Exp Tech. https://doi.org/10.1007/s40799-025-00786-5
[6]Khoshaim, A.B., Elsheikh, A.H., Moustafa, E.B., et al. (2021). Experimental investigation on laser cutting of PMMA sheets: effects of process factors on kerf characteristics. J. Mater. Res. Technol., 11, 235–246. https://doi.org/10.1016/j.jmrt.2021.01.012
[7]Caiazzo, F., Curcio, F., Daurelio, G., Minutolo, F.M.C. (2005). Laser cutting of different polymeric plastics by a CO₂ laser beam. J. Mater. Process Technol., 159(3), 279–285. https://doi.org/10.1016/j.jmatprotec.2004.02.019
[8]Rajput, A.S., Das, M., & Kapil, S. (2025). Machine-learning-based optimization of hybrid electrochemical magnetorheological finishing process to achieve nano finishing on additively manufactured biomaterial. Adv. Mater. Process Technol. https://doi.org/10.1080/2374068X.2025.2474327
[9]Kristijan, S., et al. (2025). Optimizing laser cutting of stainless steel using Latin hypercube sampling and neural networks. Opt. Laser Technol., 182, 112220. https://doi.org/10.1016/j.optlastec.2024.112220
[10]Parthiban, A., et al. (2025). Comparative study on EDM wire cutting and CO₂ laser cutting for high-precision stainless steel sheet processing. Eng. Reports, 7(5): e70201. https://doi.org/10.1002/eng2.70201
[11]Kechagias, J.D., Fountas, N.A., Ninikas, K., Vaxevanidis, N.M. (2023). Kerf geometry and surface roughness optimization in CO₂ laser processing using neural networks and genetic algorithms. J. Manuf. Mater. Process., 7(2): 77. https://doi.org/10.3390/jmmp7020077
[12]Zhu, H., et al. (2025). Postprocessing optimization for surface finishing by machine learning. In Machine Learning for Powder-Based Metal Additive Manufacturing, Elsevier, pp. 229–241. https://doi.org/10.1016/B978-0-443-22145-3.00008-6
[13]Cvijanovic, S. (2025). Artificial intelligence-driven monitoring of surface polishing by laser remelting: process stability and regime classification. J. Manuf. Processes, 137, 320–341.
[14]Cheng, Y., et al. (2025). Ultrasonic vibration-assisted laser polishing (UVLP) and establishment of roughness prediction model. Opt. Laser Technol., 186, 112714. https://doi.org/10.1016/j.optlastec.2025.112714
[15]Zhang, C., et al. (2025). Unraveling surface roughness variations in SLM-GH4169 alloy polishing using mechanistic modeling and machine learning. Mater. Today Commun., 46, 112505. https://doi.org/10.1016/j.mtcomm.2025.112505
[16]Soni, H., et al. (2025). Defect mechanisms and process optimization in laser powder bed fusion using hybrid models. J. Mater. Eng. Perform., 34(4): 1–20.
[17]Guan, Y., et al. (2025). Laser Polishing Technology. In Adv. Finishing Technol. for High Performance Manufacturing, Springer Nature Singapore. https://doi.org/10.1007/978-981-96-1900-9_10
[18]Sinico, M., Witvrouw, A., & Dewulf, W. (2025). Improving surface quality of maraging steel parts via optimized laser remelting. J. Manuf. Processes, 138, 38–49. https://doi.org/10.1016/j.jmapro.2025.02.014
[19]Mechali, A., et al. (2025). Surface and post-processing characteristics of maraging steel fabricated via SLM. Int. J. Adv. Manuf. Technol., 1–16.
[20]Reddy, B.V.S., et al. (2025). Performance evaluation of machine learning techniques in surface roughness prediction for 3D printed structures. J. Manuf. Processes, 137, 320–341. https://doi.org/10.1016/j.jmapro.2025.01.082
[21]Gladstone, R.J., et al. (2025). FO-PINN: A first-order formulation for physics-informed neural networks. Eng. Anal. Bound. Elem., 174, 106161. https://doi.org/10.1016/j.enganabound.2025.106161
[22]Berardi, M., Difonzo, F.V., & Icardi, M. (2025). Inverse physics-informed neural networks for transport models in porous materials. Comput. Methods Appl. Mech. Eng., 435, 117628. https://doi.org/10.1016/j.cma.2024.117628
[23]Jalili, D., & Mahmoudi, Y. (2025). Physics-informed neural networks for two-phase film boiling heat transfer. Int. J. Heat Mass Transf., 241, 126680. https://doi.org/10.1016/j.ijheatmasstransfer.2025.126680
[24]Wijerathne, A.D.H.T., et al. (2025). Recent advances in food drying modeling: empirical to multiscale physics-informed neural networks. Compr. Rev. Food Sci. Food Saf., 24(3): e70194. https://doi.org/10.1111/1541-4337.70194
[25]Ye, B., et al. (2025). iMLGAM: Integrated machine learning and genetic algorithm-driven multiomics analysis for prediction. Imeta, 4(2): e70011. https://doi.org/10.1002/imt2.70011
[26]Alexakis, K., et al. (2025). Genetic algorithm-based multi-objective optimisation for energy-efficient systems: a systematic review. Energy Build., 328, 115216. https://doi.org/10.1016/j.enbuild.2024.115216
[27]Zhang, W., et al. (2025). Multi-physics coupling model parameter identification using data-driven methods and genetic algorithms. Energy, 314, 134120. https://doi.org/10.1016/j.energy.2024.134120
[28]Adabbo, G., et al. (2025). A multi-objective optimization framework through genetic algorithm for hyperthermia-mediated drug delivery. Comput. Biol. Med., 189, 109895. https://doi.org/10.1016/j.compbiomed.2025.109895
[29]Karkadakattil, A. (2025). AI-driven prediction of surface roughness in laser-polished LPBF Ti-6Al-4V: a sustainable proof-of-concept. Austral. J. Multi-Disciplinary Eng., 1–13. https://doi.org/10.1080/14488388.2025.2570030
[30]Donachie, M.J. (2002). Superalloys: A Technical Guide. ASM International. https://doi.org/10.31399/asm.tb.stg2.9781627082679
[31]Davim, J.P. (2008). Machining: Fundamentals and Recent Advances. Springer.
[32]Groover, M.P. (2013). Fundamentals of Modern Manufacturing: Materials, Processes, and Systems. John Wiley & Sons.
[33]Davim, J.P. (2012). Statistical and Computational Techniques in Manufacturing. Springer. https://doi.org/10.1007/978-3-642-25859-6
[34]Li, T., et al. (2025). Experimental study on laser cutting of stainless steel hexagonal tube of reactor assemblies. Nucl. Eng. Des., 432, 113788. https://doi.org/10.1016/j.nucengdes.2024.113788
[35]Zabon, A.H., Abbas, T.F., & Bedan, A.S. (2025). Enhancing laser cutting quality of stainless steel 201 using multi-factor design. Adv. Sci. Technol. Res. J., 19(7): 458–470. https://doi.org/10.12913/22998624/204716
[36]Aziz, U., et al. (2025). Optimization of additively manufactured 316/316L stainless steel process parameters and post-processing strategies. Materials, 18(12): 2870. https://doi.org/10.3390/ma18122870
[37]Vishnulal, R., Govindan, P., & Vipindas, M. (2020). Laser machining of polymer materials: process challenges and strategies. In Springer ICETE Proc., pp. 653–661. https://doi.org/10.1007/978-3-030-24314-2_77
[38]Karkadakattil, A. Geometry aware laser polishing of LPBF AlSi10Mg defence components with physics inspired neural network based surface roughness prediction. Discov Mechanical Engineering 4, 52 (2025). https://doi.org/10.1007/s44245-025-00144-0
[39]Mousavian, R.T., et al. (2020). Development of BMG-B2 nanocomposite structure during laser surface processing. Appl. Surf. Sci., 505, 144535. https://doi.org/10.1016/j.apsusc.2019.144535
[40]Patel, G.C.M., et al. (2021). Optimization of EDM parameters using hybrid Taguchi-based PCA and CRITIC approaches. Metals, 11(3): 10007.