This page summarizes recent research efforts (2023 - 2026). Past research can be found in Google Scholar.

Theme 1 - In-situ Process Monitoring

We leverage advances in AI/ML to enable real-time monitoring of manufacturing processes by over coming the limitations of data transfer bandwidth and/or sensors themselves. The recent projects are shown below. Sponsors/collaborators: Los Alamos National Lab 
Neuromorphic imaging and sparse sensing:  This work uses event-based imager and sparse sensing techniques to develop a low-latency, low-data melt pool monitoring framework, enabling a truly real-time application. read more:
  • Heydari, D. Mascarenas, CC Shih, and B. Tai*, 2026, “Toward memory-efficient melt pool monitoring: a classification framework using event-based imaging and sparse sensing technique,” Journal of Intelligent Manufacturing, 1-19. https://doi.org/10.1007/s10845-026-02808-4
 
High-frequency dynamic force measurement: This work uses neuro-operator learning to map acceleration data to dynamic cutting forces in machining, overcoming the limitations of traditional sensor size and natural frequency. Read more:
  • CC. Shih and B. Tai*, 2025, “Dynamic Cutting Force Estimation via Fourier Neural Operator with Inferred Machine Tool Dynamics: A Proof of Concept,” Journal of Manufacturing Science and Engineering, 147(8), 081003. https://doi.org/10.1115/1.4068643

Theme 2 - Advanced Packaging

We apply our expertise in thermal-mechanical modeling and inverse methods to tackle challenges in multi-scale, heterogeneous 3D features in advanced semiconductor packaging. The following projects are relevant but not directly related to the theme due to confidentiality. Sponsors/collaborators: TSMC (confidential projects), NYCU
AI-enabled surrogate modeling: This work introduces a PINN-based framework to replace CFD, demonstrating the ability of AI to surrogate a numerical model.  Read more:
  • CC. Shih, P.T. Lin, D. Shektybayeva, U. Braga-neto, and B. Tai*, 2026, “Physics-Informed Hypernetworks for Generalized Obstacle-Flow Prediction with Varying Shapes and Locations,” Computer Methods in Applied Mechanics and Engineering, in press.
 
Material properties: This work uses an AI-powered inverse method to determine the material's constitutive model, which is otherwise difficult to measure. Read more:
  • PT. Lin, H. Chawla, and B. Tai*, 2026, “Learning Johnson–Cook Parameters from Chip Formation and Cutting Forces Using a Multimodal Machine Learning Framework,” Journal of Manufacturing Science and Engineering, 148 (5), 051008 https://doi.org/10.1115/1.4071283

Theme 3 - Custom-AM

This has been a long focus of the lab. We develop additive manufacturing processes (mostly polymers) for special designs, materials, or enhanced properties. Some recent work is shown below. Sponsors/collaborators: TAMUQ/Qatar Foundation, internal funds
Hybrid AM 1(FDM + machining): This work aims to develop a machining + FDM platform to enable high-precision, fine-finish polymer AM, especially for high-performance composite filaments. Read more:
  • Nigam and B. Tai*, 2024, “Effects of In-Process Surface Finishing on Part Strength in Polymer Material Extrusion Additive Manufacturing,” Additive Manufacturing, v 80, p. 103960. https://doi.org/10.1016/j.addma.2024.103960
  • Nigam and B. Tai*, 2023, “Surface Characterization of Three-Dimensional Printed Fiber-Reinforced Polymer Following an In-Process Mechanical–Chemical Finishing Method, Journal of Manufacturing Science and Engineering, v145, 081002, https://doi.org/10.1115/1.4062146
 
Hybrid AM 2 (FDM + Injection molding): This work aims to develop an FDM + injection molding platform to enhance part integrity and strength by overcoming weak interfaces. This is a collaboration with TAMUQ (Qatar). Read more:
  • J. Katbeh, et al. 2025, "Toward a hybrid material extrusion technique: exploring the benefits of integrating fused deposition modeling and material injection." Progress in Additive Manufacturing 10, 11343–11360. https://doi.org/10.1007/s40964-025-01290-3
 

Theme 4 - Machining

This has been the root of the lab. We perform experimental studies and simulations for analyzing a variety of machining processes of both engineering and non-engineering materials. Sponsors/collaborators: KCNSC (DOE-NNSA), GTO (DOE-EERE), NSF
AM metal alloys: This work experimentally examines the machinability of various AM alloys as compared to their wrought or case counterparts. Read more:
  • Randolph, R. Zvanut and B. Tai*, 2024, "Microstructure analysis and machinability of additively manufactured A205 aluminum with heat treatments," Journal of Manufacturing Processes, 127, pp. 51-61. https://doi.org/10.1016/j.jmapro.2024.07.111
  • Raval, A. Kazi, O. Randolph, X. Guo, R. Zvanut, C. Lee, B. Tai*, 2023, “Machinability comparison of additively manufactured and traditionally wrought Ti-6Al-4V alloys using single-point cutting”, Journal of Manufacturing Processes, v 95, pp. 539-549 https://doi.org/10.1016/j.jmapro.2023.03.041
Plasma-accelerated rock drilling: This is a collaborative effort with Dr. Antao and Dr. Staack to develop a plasma-integrated rock drill that enables pre-cracking during drilling to increase the rate of penetration (ROP). Read more:
  • Akhter, X. Tang, J. Mallams, Y. Kao, A. Kazi, S. Kumar, D.S. Antao, B. L. Tai, D. Staack*, 2025, “Shockwave and plasma assisted rock cracking for geothermal drilling,” Renewable Energy. https://doi.org/10.1016/j.renene.2025.122351
  • Kazi, M. Akhter, D. Antao, D. Staack, and B. Tai*, 2024, ""Numerical Investigation of the Effect of Pre-induced Cracks on Hard Rock Cutting Using Finite Element Analysis " Rock Mechanics and Rock Engineering https://doi.org/10.1007/s00603-024-03990-8

Theme 5- Surgical Simulation and Training

We design and implement surgical simulation systems for orthopaedic surgeons (particularly for bone drilling) and develop data-driven and AI tools for quantitative feedback in aid of training. Sponsors/collaborators: ABOS, UT Health at Houston, Penn State
Data-driven bone drilling platform: Working with orthopaedic surgeons, this project develops a hardware platform and analysis method for resident training and assessment. Read more:
  • Nigam, JF Kellam, CG Ambrose, and B. Tai*, 2025, “Impact of Quantitative Feedback on Motor Skill Acquisition in Bone Drilling using Radar Plots,” Journal of Orthoapedic Trauma, 39, S21-S26. DOI:10.5281/zenodo.17411489
  • Nigam, J. Kellam, C. Ambrose, and B. Tai*, 2024, “A Data-Driven Methodology to Comprehensively Assess Bone Drilling Using Radar Plots,” Journal of Bone & Joint Surgery Open Access 9(1) https://dx.doi.org/10.2106/JBJS.OA.23.00069
 
AI-supervised Scoring System: This work aims to develop an interpretable AI/ML framework to quantitatively classify skill levels using both hidden and user-defined features.
  • Y. Chen, D. Shektybayeva, et al., 2026, "An AI-based Framework for Automatic Skill Level Assessment and Scoring in Surgical Bone Drilling Training," manuscript in review.