{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,14]],"date-time":"2026-09-14T11:05:01Z","timestamp":1789383901852,"version":"build-2803163510"},"reference-count":50,"publisher":"MDPI AG","issue":"18","license":[{"start":{"date-parts":[[2026,9,14]],"date-time":"2026-09-14T00:00:00Z","timestamp":1789344000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Achieving Zero-Defect Manufacturing (ZDM) in precision micro-injection molding requires continuous monitoring of non-linear interactions among continuous thermodynamic and kinetic sensor data (e.g., melt temperature, injection pressure) and discrete equipment states. The high-throughput production of optical lenses operates under stringent physical boundaries, limiting spherical power deviations to a \u00b10.25 Diopters (D) threshold. This study proposes a sensor-driven digital twin framework for virtual metrology (VM) and prescriptive decision support. The proposed hybrid framework operationalizes Knowledge-Informed Machine Learning (KIML) by imposing physical constraints during evolutionary optimization and auditing the learned internal interactions via a dual-level Explainable AI (XAI) protocol. Utilizing 175,089 sensor logs, a continuous surrogate is constructed via the Feature Tokenizer Transformer (FT-Transformer). Under randomized 5-fold cross-validation, the surrogate achieves a Root Mean Squared Error (RMSE) of 0.4211 D and a Coefficient of Determination (R2) of 0.97, demonstrating a 16.8% error reduction over the 1D-CNN baseline. To rigorously evaluate inter-machine generalization and mitigate batch-level data leakage, an equipment-isolated Leave-One-Machine-Out (LOMO) protocol confirms structural robustness with an R2 of 0.8820. Furthermore, residual distribution analysis demonstrates that 76.19% of the validation samples actively fall within the \u00b10.25 D physical tolerance. Cross-verifying intrinsic Multi-Head Self-Attention (MHSA) weights against a global proxy statistically captures the underlying associative affinities between hardware and continuous sensor metrics. Integrating this differentiable surrogate with a real-valued Genetic Algorithm (GA) enables the autonomous generation of optimized process recipes, achieving an algorithmic convergence error below 0.001 D within the continuous latent space. While future physical validation remains necessary, this framework establishes a transparent, auditable foundation for prescriptive smart manufacturing.<\/jats:p>","DOI":"10.3390\/s26185807","type":"journal-article","created":{"date-parts":[[2026,9,14]],"date-time":"2026-09-14T10:19:41Z","timestamp":1789381181000},"page":"5807","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Dual-Level XAI-Guided Digital Twin Framework for Prescriptive Decision Making in Sensor-Driven Manufacturing"],"prefix":"10.3390","volume":"26","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-1058-3257","authenticated-orcid":false,"given":"Joonyong","family":"Park","sequence":"first","affiliation":[{"name":"Department of Global Business, Busan International College (BIC), Tongmyong University, Busan 48520, Republic of Korea"},{"name":"DXIAI (Digital Transformation Industrial AI), Institute of ABH Inc., Busan 48059, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,9,14]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"111","DOI":"10.1007\/s40684-016-0015-5","article-title":"Smart manufacturing: Past research, present findings, and future directions","volume":"3","author":"Kang","year":"2016","journal-title":"Int. J. Precis. Eng. Manuf.-Green Technol."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1080\/00207543.2019.1605228","article-title":"Zero-defect manufacturing: State-of-the-art review, shortcomings and future directions in research","volume":"58","author":"Psarommatis","year":"2020","journal-title":"Int. J. Prod. Res."},{"key":"ref_3","first-page":"234","article-title":"Data-driven smart manufacturing: From defect prediction to prescriptive analytics","volume":"57","author":"Gao","year":"2020","journal-title":"J. Manuf. Syst."},{"key":"ref_4","first-page":"145","article-title":"Simulation and experimental study on the optical properties of injection-molded optical lenses","volume":"48","author":"Chen","year":"2008","journal-title":"Polym. Eng. Sci."},{"key":"ref_5","first-page":"281","article-title":"Slim mobile lens design using a hybrid refractive\/diffractive lens","volume":"31","author":"Park","year":"2020","journal-title":"Korean J. Opt. Photonics"},{"key":"ref_6","unstructured":"(2017). Ophthalmic Optics\u2014Contact Lenses\u2014Part 2: Tolerances (Standard No. ISO 18369-2:2017)."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"43","DOI":"10.7469\/JKSQM.2024.52.1.43","article-title":"Methodology for variable optimization in injection molding process","volume":"52","author":"Jung","year":"2024","journal-title":"J. Korean Soc. Qual. Manag."},{"key":"ref_8","first-page":"123","article-title":"Optimization of injection molding process parameters for minimizing warpage using the Taguchi method","volume":"15","author":"Park","year":"2014","journal-title":"Int. J. Precis. Eng. Manuf."},{"key":"ref_9","first-page":"45","article-title":"Benefit evaluation of virtual metrology","volume":"31","author":"Cheng","year":"2018","journal-title":"IEEE Trans. Semicond. Manuf."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"3211","DOI":"10.1007\/s00170-023-12726-x","article-title":"Virtual metrology for enabling zero-defect manufacturing: A review and prospects","volume":"130","author":"Zhang","year":"2024","journal-title":"Int. J. Adv. Manuf. Technol."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"144","DOI":"10.1016\/j.jmsy.2018.01.003","article-title":"Deep learning for smart manufacturing: Methods and applications","volume":"48","author":"Wang","year":"2018","journal-title":"J. Manuf. Syst."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"7499","DOI":"10.1109\/TNNLS.2022.3229161","article-title":"Deep neural networks and tabular data: A survey","volume":"35","author":"Borisov","year":"2024","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."},{"key":"ref_13","first-page":"207","article-title":"Development of quality defect prediction model for injection molding products using machine learning","volume":"18","author":"Han","year":"2025","journal-title":"J. Korea Inst. Inf. Electron. Commun. Technol."},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Chen, T., and Guestrin, C. (2016, January 13\u201317). XGBoost: A scalable tree boosting system. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, USA.","DOI":"10.1145\/2939672.2939785"},{"key":"ref_15","first-page":"3146","article-title":"LightGBM: A highly efficient gradient boosting decision tree","volume":"30","author":"Ke","year":"2017","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_16","first-page":"18932","article-title":"Revisiting deep learning models for tabular data","volume":"34","author":"Gorishniy","year":"2021","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_17","first-page":"2104","article-title":"Transtab: Learning transferable tabular transformers across tables","volume":"35","author":"Wang","year":"2022","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"He, K., Zhang, X., Ren, S., and Sun, J. (2016, January 27\u201330). Deep residual learning for image recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA.","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref_19","first-page":"23928","article-title":"Well-tuned simple nets excel on tabular datasets","volume":"34","author":"Kadra","year":"2021","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_20","first-page":"6679","article-title":"TabNet: Attentive interpretable tabular learning","volume":"35","author":"Arik","year":"2021","journal-title":"Proc. AAAI Conf. Artif. Intell."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"507","DOI":"10.52202\/068431-0037","article-title":"Why do tree-based models still outperform deep learning on typical tabular data?","volume":"35","author":"Grinsztajn","year":"2022","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_22","first-page":"76336","article-title":"When do neural nets outperform boosted trees on tabular data?","volume":"36","author":"McElfresh","year":"2024","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"5134","DOI":"10.1109\/TCYB.2026.3691881","article-title":"HoStB-DVGNN: A Flotation Fault Recognition Method Using Higher Order Spatial-Temporal Block and Dual-Stream Variational GNN","volume":"56","author":"Fan","year":"2026","journal-title":"IEEE Trans. Cybern."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"4860","DOI":"10.1109\/TII.2026.3668271","article-title":"Flotation Fault Trace Recognition Using Dynamic Edge Weight-Based Cross-Cell Interaction Graph Transformer and Joint Task Learning","volume":"22","author":"Fan","year":"2026","journal-title":"IEEE Trans. Ind. Inform."},{"key":"ref_25","unstructured":"Willard, J., Jia, X., Xu, S., Steinbach, M., and Kumar, V. (2020). Integrating physics-based modeling with machine learning: A survey. arXiv."},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"422","DOI":"10.1038\/s42254-021-00314-5","article-title":"Physics-informed machine learning","volume":"3","author":"Karniadakis","year":"2021","journal-title":"Nat. Rev. Phys."},{"key":"ref_27","first-page":"5998","article-title":"Attention is all you need","volume":"30","author":"Vaswani","year":"2017","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Holland, J.H. (1992). Adaptation in Natural and Artificial Systems, MIT Press.","DOI":"10.7551\/mitpress\/1090.001.0001"},{"key":"ref_29","unstructured":"Grieves, M. (2014). Digital Twin: Mitigating Unpredictable, Unsustainable Capability Risks, Executive Brief."},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Tao, F., Zhang, M., and Nee, A.Y.C. (2019). Digital Twin Driven Smart Manufacturing, Academic Press.","DOI":"10.1016\/B978-0-12-817630-6.00010-2"},{"key":"ref_31","unstructured":"Ministry of SMEs and Startups, and KAIST (2026, August 12). Production Process Optimization AI Dataset; Korea AI Manufacturing Platform (KAMP). Available online: https:\/\/www.kamp-ai.kr\/."},{"key":"ref_32","unstructured":"Gorishniy, Y., Rubachev, I., and Babenko, A. (2024). TabM: Advancing tabular deep learning. arXiv."},{"key":"ref_33","unstructured":"Huang, X., Khetan, A., Cvitkovic, M., and Karnin, Z. (2020). TabTransformer: Tabular data modeling using contextual embeddings. arXiv."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"61","DOI":"10.1016\/j.swevo.2011.05.001","article-title":"Surrogate-assisted evolutionary computation: Recent advances and future challenges","volume":"1","author":"Jin","year":"2011","journal-title":"Swarm Evol. Comput."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"442","DOI":"10.1109\/TEVC.2018.2869001","article-title":"Data-driven evolutionary optimization: An overview and case studies","volume":"23","author":"Jin","year":"2018","journal-title":"IEEE Trans. Evol. Comput."},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"7775","DOI":"10.1038\/s41598-024-56699-7","article-title":"A GAN-based genetic algorithm for solving the 3D bin packing problem","volume":"14","author":"Zhang","year":"2024","journal-title":"Sci. Rep."},{"key":"ref_37","first-page":"210","article-title":"Physics-constrained evolutionary algorithms for inverse design in cyber-physical production systems","volume":"68","author":"Liu","year":"2023","journal-title":"J. Manuf. Syst."},{"key":"ref_38","first-page":"1450","article-title":"Surrogate-assisted evolutionary algorithms for mixed-integer manufacturing optimization","volume":"54","author":"Chen","year":"2024","journal-title":"IEEE Trans. Cybern."},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"40","DOI":"10.1214\/09-SS054","article-title":"A survey of cross-validation procedures for model selection","volume":"4","author":"Arlot","year":"2010","journal-title":"Stat. Surv."},{"key":"ref_40","first-page":"45","article-title":"A new typology design of performance metrics to measure errors in machine learning regression algorithms","volume":"14","author":"Botchkarev","year":"2019","journal-title":"Interdiscip. J. Inf. Knowl. Manag."},{"key":"ref_41","first-page":"93","article-title":"A survey of methods for explaining black box models","volume":"51","author":"Guidotti","year":"2018","journal-title":"ACM Comput. Surv."},{"key":"ref_42","first-page":"4768","article-title":"A unified approach to interpreting model predictions","volume":"30","author":"Lundberg","year":"2017","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_43","doi-asserted-by":"crossref","unstructured":"Ribeiro, M.T., Singh, S., and Guestrin, C. (2016, January 13\u201317). \u201cWhy should I trust you?\u201d Explaining the predictions of any classifier. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, USA.","DOI":"10.1145\/2939672.2939778"},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"31","DOI":"10.1145\/3236386.3241340","article-title":"The mythos of model interpretability","volume":"16","author":"Lipton","year":"2018","journal-title":"ACM Queue"},{"key":"ref_45","doi-asserted-by":"crossref","first-page":"31150","DOI":"10.1038\/s41598-024-82501-9","article-title":"Explainable AI improves task performance in human-AI collaboration","volume":"14","author":"Senoner","year":"2024","journal-title":"Sci. Rep."},{"key":"ref_46","doi-asserted-by":"crossref","unstructured":"Wiegreffe, S., and Pinter, Y. (2019, January 3\u20137). Attention is not not explanation. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing (EMNLP), Hong Kong, China.","DOI":"10.18653\/v1\/D19-1002"},{"key":"ref_47","doi-asserted-by":"crossref","first-page":"508","DOI":"10.1080\/00207543.2017.1351644","article-title":"Smart manufacturing","volume":"56","author":"Kusiak","year":"2018","journal-title":"Int. J. Prod. Res."},{"key":"ref_48","doi-asserted-by":"crossref","first-page":"44","DOI":"10.1145\/2523813","article-title":"A survey on concept drift adaptation","volume":"46","author":"Gama","year":"2014","journal-title":"ACM Comput. Surv."},{"key":"ref_49","unstructured":"Hinton, G., Vinyals, O., and Dean, J. (2015). Distilling the knowledge in a neural network. arXiv."},{"key":"ref_50","doi-asserted-by":"crossref","first-page":"182","DOI":"10.1109\/4235.996017","article-title":"A fast and elitist multiobjective genetic algorithm: NSGA-II","volume":"6","author":"Deb","year":"2002","journal-title":"IEEE Trans. Evol. Comput."}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/18\/5807\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,14]],"date-time":"2026-09-14T10:23:41Z","timestamp":1789381421000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/18\/5807"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,14]]},"references-count":50,"journal-issue":{"issue":"18","published-online":{"date-parts":[[2026,9]]}},"alternative-id":["s26185807"],"URL":"https:\/\/doi.org\/10.3390\/s26185807","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,14]]}}}