Computer-aided design (CAD) is vital to modern manufacturing, yet model creation remains labor-intensive and expertise-heavy. To enable non-experts to translate intuitive design intent into manufacturable artifacts, recent large language models-based text-to-CAD efforts focus on command sequences or script-based formats like CadQuery. However, these formats are kernel-dependent and lack universality for manufacturing. In contrast, the Standard for the Exchange of Product Data (STEP, ISO 10303) file is a widely adopted, neutral boundary representation (B-rep) format directly compatible with manufacturing, but its graph-structured, cross-referenced nature poses unique challenges for auto-regressive LLMs. To address this, we curate a dataset of 40K STEP-caption pairs and introduce novel preprocessing tailored for the graph-structured format of STEP, including a depth-first search-based reserialization that linearizes cross-references while preserving locality and chain-of-thought(CoT)-style structural annotations that guide global coherence. We integrate retrieval-augmented generation to ground predictions in relevant examples for supervised fine-tuning, and refine generation quality through reinforcement learning with a specific Chamfer Distance-based geometric reward. Experiments demonstrate consistent gains of our STEP-LLM in geometric fidelity over the Text2CAD baseline, with improvements arising from multiple stages of our framework: the RAG module substantially enhances completeness and renderability, the DFS-based reserialization strengthens overall accuracy, and the RL further reduces geometric discrepancy. Both metrics and visual comparisons confirm that STEP-LLM generates shapes with higher fidelity than Text2CAD. These results show the feasibility of LLM-driven STEP model generation from natural language, showing its potential to democratize CAD design for manufacturing.

Digital Twin (DT) technologies are transforming manufacturing by enabling real-time prediction, monitoring, and control of complex processes. Yet, applying DT to deformation-based metal forming remains challenging because of the strongly coupled spatial-temporal behavior and the nonlinear relationship between toolpath and material response. For instance, sheet-metal forming by the English wheel, a highly flexible but artisan-dependent process, still lacks digital counterparts that can autonomously plan and adapt forming strategies. This study presents an adaptive DT framework that integrates Proper Orthogonal Decomposition (POD) for physics-aware dimensionality reduction with a Koopman operator for representing nonlinear system in a linear lifted space for the real-time decision-making via model predictive control (MPC). To accommodate evolving process conditions or material states, an online Recursive Least Squares (RLS) algorithm is introduced to update the operator coefficients in real time, enabling continuous adaptation of the DT model as new deformation data become available. The proposed framework is experimentally demonstrated on a robotic English Wheel sheet metal forming system, where deformation fields are measured and modeled under varying toolpaths. Results show that the adaptive DT is capable of controlling the forming process to achieve the given target shape by effectively capturing non-stationary process behaviors. Beyond this case study, the proposed framework establishes a generalizable approach for interpretable, adaptive, and computationally-efficient DT of nonlinear manufacturing systems, bridging reduced-order physics representations with data-driven adaptability to support autonomous process control and optimization.

Laser-based directed energy deposition (DED) with in situ powder blending is used to fabricate stepped-composition alloy couples, enabling rapid property screening of large composition spaces. Composition grading occurs between binary systems of four commercial superalloys: Ni-based Inconel 625 (IN625), Co-based Haynes 25 (HA25), Fe-based stainless steel 316L, and Ni-based Nimonic 80A (N80A). The microstructure and ambient temperature microhardness of these graded couples are analyzed in the as-fabricated state and after heat-treatment (homogenized + inter-diffused 1150 °C / 72 h, aged 800 °C / 24 h). A linear rule-of-mixtures (ROM) model is applied to composition-correlated microhardness to assess (i) the predictive power of ROM between solid-solution strengthened alloys (IN625, HA25, and 316L), and (ii) deviations in ROM as a screening mechanism for precipitation-hardening in Al- and Ti-containing compositions (based on N80A). Single-composition specimens of the four basis alloys and three near-50:50 blends (wt. %) between IN625, HA25, and 316L are subjected to creep in the temperature range 700–850°C. Creep pre-exponential constant A, stress exponent n, and apparent activation energy Qa are extracted for these select compositions; linear ROM is then applied to individual creep coefficients between basis alloys for comparison. Resulting strain rate predictions suggest that blended alloy creep behavior is sufficiently predicted via ROM within a 90% predictive interval of ×10^(±0.39) (2.5×), which is remarkably accurate with respect to the predictive power of the creep power law (×10^(±0.21), 1.6×).

We present a fully differentiable, GPU-accelerated finite element framework JAX-FEM-ANISO, for forward simulation and inverse parameter identification of finite-strain anisotropic plasticity. Built on JAX-FEM, the framework exploits modern accelerator architectures by parallelizing the three major computational bottlenecks in nonlinear FEM: elemental weak-form and tangent-stiffness evaluation, global sparse matrix assembly, and sparse linear solution. For a large-scale forward problem with 3 million degrees of freedom, JAX-FEM-ANISO on a single NVIDIA H100 GPU achieves up to 9.4× speed-up over a 24-core CPU Abaqus baseline. Automatic differentiation is applied through the constitutive update and solver workflow, providing consistent Jacobians for complex constitutive models without manual derivation and accurate gradients for PDE-constrained inverse analysis. Compared with finite differences, the JAX-AD gradients avoid step-size sensitivity and provide the required sensitivities at substantially lower computational cost. For inverse characterization, we combine information-rich, topology-optimized heterogeneous specimens with full-field displacement data to identify advanced constitutive model parameters from a single test, replacing what would otherwise require many conventional experiments. We demonstrate accurate recovery of anisotropic yield and hardening parameters in progressively challenging settings, including uniform and spatially varying material properties. The resulting AD-based formulation enables efficient optimization in high-dimensional parameter spaces where finite-difference approaches are computationally infeasible. These results establish differentiable, GPU-accelerated FEM as a practical high-throughput engine for simulation, characterization, and optimization workflows in advanced manufacturing.

Accurate and efficient determination of crystal plasticity (CP) material parameters is essential for predictive simulations that link microstructures, manufacturing processes, and material properties. This study presents a graphical processing unit (GPU)-accelerated pipeline for calibrating CP material parameters, integrating automatic differentiation (AD)-based sensitivities with gradient-based optimization, built upon our open-source jax-cpfem package. This method eliminates reliance on finite differences in gradient-based approaches while improving efficiency over gradient-free optimization. The effectiveness of the pipeline is demonstrated through five case studies covering various crystal structures and boundary conditions. First, the AD-based sensitivity analysis achieves over 10 × speedup compared to finite difference while maintaining accuracy for complex, nonlinear constitutive laws. Second, a comprehensive analysis of initial starting points on gradient-based optimization demonstrates that using appropriate bounds mitigates potential issues. Across both single-crystal and polycrystalline cases calibrating six material parameters, our pipeline requires approximately 7 × fewer iterations and achieves 3 × higher efficiency over popular gradient-free methods like Bayesian optimization, regardless of geometry complexity. Furthermore, the successful calibration of 12 parameters in a dual-phase steel model highlights the capability of the pipeline to handle high-dimensional optimization problems, which is challenging for gradient-free optimization. Finally, the robustness of our pipeline is validated using noisy synthetic data and experimental tensile data for wrought IN625 over a finite strain range. These results illustrate the applicability of our pipeline to real-world scenarios and its potential for high-dimensional optimization and promising applications in integrated computational materials engineering workflows.

High-temperature creep characterization of structural alloys traditionally relies on serial uniaxial tests, which are highly inefficient for exploring the large search space of alloy compositions and for material discovery. Here, we introduce a machine-learning-assisted, high-throughput framework for creep law identification based on a dimple array bulge instrument (DABI) configuration, which enables parallel creep testing of 25 dimples, each fabricated from a different alloy, in a single experiment. Full-field surface displacements of dimples undergoing time-dependent creep-induced bulging under inert gas pressure are measured by 3D digital image correlation. We train a recurrent neural network as a surrogate model, mapping creep parameters and loading conditions to the time-dependent deformation response of DABI. Coupling this surrogate with a particle swarm optimization scheme enables rapid and global inverse identification with sparsity regularization of creep parameters from experimental displacement–time histories. In addition, we propose a phenomenological creep law with a time-dependent stress exponent that captures the sigmoidal primary creep observed in wrought INCONEL 625 and extracts its temperature dependence from DABI tests at multiple temperatures. Furthermore, we employ a general creep law combining several conventional forms with regularized inversion to identify the creep laws for 47 additional Fe-, Ni-, and Co-rich alloys and to automatically select the dominant functional form for each alloy. This workflow, combined with the DABI experiment, provides a quantitative, high-throughput creep characterization platform that is compatible with data mining, composition–property modeling, and nonlinear structural optimization with creep behavior across a large alloy design space.

We present JAX-PF, an open-source, GPU-accelerated, and differentiable Phase Field (PF) software package, supporting both explicit and implicit time stepping schemes. Leveraging the modern computing architecture JAX, JAX-PF achieves high performance through array programming and GPU acceleration, delivering 5x speedup over PRISMS-PF with MPI (24 CPU cores) for systems with 4.19 million degrees of freedom using explicit schemes, and scaling efficiently with implicit schemes for large-size problems. Furthermore, a key feature of JAX-PF is automatic differentiation (AD), eliminating manual derivations of free-energy functionals and Jacobians. Beyond forward simulations, JAX-PF demonstrates its potential in inverse design by providing sensitivities for gradient-based optimization. We demonstrate, for the first time, the calibration of PF material parameters using AD-based sensitivities, highlighting its capability for high-dimensional inverse problems. By combining efficiency, flexibility, and full differentiability, JAX-PF offers a fast, practical, and integrated tool for forward simulation and inverse design, advancing co-designing of material and manufacturing processes and supporting the goals of the Materials Genome Initiative.

Creep is a primary life-limiting mechanism for metallic components operating at high temperature, producing permanent deformation under sustained loads even when stresses remain below yield. The design of structures to minimize this deformation is critical to extending the service life of components. Incorporating creep into topology optimization (TO) remains open because the response is nonlinear, history-dependent, and thermomechanically coupled, and prior work often relies on linear viscoelastic models, which do not capture the behavior of metals at high temperatures. To bridge this gap, we introduce a differentiable thermo-structural TO framework. The approach considers creep deformation using the Norton model and leverages JAX's automatic differentiation to perform adjoint sensitivity analysis, enabling efficient gradient-based optimization. The transient material response is solved via a backward Euler scheme over a prescribed service life. Our objective is to minimize creep deformation subject to a volume constraint. We first demonstrate the framework on canonical two-dimensional benchmarks, showing that the proposed formulation significantly reduces permanent deformation compared to designs optimized solely for elastic stiffness. We then pose, as a challenge problem, the compositional design of a three-dimensional graded material turbine blade in which the local mixture of two candidate alloys is optimized. This challenge problem exercises the full capability of the framework, including transient nonlinear creep, coupled thermal loading, three-dimensional geometry, and gradient-based multi-material design, highlighting the need for creep-aware design in high-temperature applications.

Reliable in-situ detection of keyhole collapse in Laser Powder Bed Fusion (LPBF) remains challenging due to the transient nature of the melt pool and the inherent stochasticity of the process. This study presents a physically informed, machine learning-based framework that identifies collapse events directly from Thermal Energy Density (TED) obtained from coaxial photodiodes with a sampling rate of 200 kHz. Power Spectral Density (PSD) analysis of the TED signal revealed that low-frequency regions are dominated by pink noise and high-frequency regions by white noise, with bulk melt pool motion and collapse dynamics confined to the intermediate band. Accordingly, a 0.25-30 kHz band-pass filter was applied to the TED signal to isolate the melt pool dynamics. Statistical and frequency-domain features from the filtered TED are used to train a Random Forest anomaly detection model using ground truth data from the Advanced Photon Source, where high-speed operando X-ray imaging verified collapse events. Applied to full build TED data from a commercial DMG MORI LASERTEC 12 SLM (LPBF) machine, the framework generalizes within the same material system and sensing modality without retraining, preserving a strong correlation between predicted anomalies and part-level CT-measured porosity. By integrating physics-based filtering, feature-driven learning, and adaptive thresholding, this method provides a scalable and interpretable foundation for real-time LPBF defect detection and monitoring.

Abnormal martensitic microstructures have been observed in additively manufactured (AMed) maraging steels. We find that AMed martensite exhibits a highly biased selection of variants, predominantly concentrated on a few variants within a single prior austenite grain. Variant selection follows orientation-dependent trends, with restricted variants exhibiting Bain axes closely aligned with specific coordinate directions. Through the thermomechanical finite element analysis, thermal strains during the deposition process are computed, and it is revealed that the tensile residual thermal strain was dominant along a specific orientation. Moreover, the orientation along which the tensile strain was applied is consistent with the Bain axis of the variants that exhibited the restricted selection behavior. We conclude that the direction-dependent distribution of thermal strain during additive manufacturing suppresses the martensitic transformation when the strain is applied in tension. This finding demonstrates potential pathways for intelligent process design, enabling spatial control of microstructure and properties for specific applications.

Stitch reinforcements have an increasing role in the design and customization of fiber reinforced polymers. Commonly known for their uses in non-crimp, stitch bonded, and 3-D textiles, innovations have expanded the role of stitches for optimization of forming behavior, improving bonding in composites, and customized textile sensors. Recently, the novel implementation of embroidery stitching for manufacturing Kirigami-based deployable structures has inspired new needs for the fundamental understanding of the effect of stitching on the resulting mechanical behavior. Embroidery stitch machines create a lockstitch: a stitch consisting of upper and lower threads intertwined to “lock” together. Despite the advancements in their manufacturing technology, stitches are under investigated in their own ability to augment composite textile behavior through design and material hybridization. While efforts have been made to model warp stitching and through thickness stitching of 3-D textiles, very little has been researched into the mechanical effect of stitch pattern, processing, material, and methods to characterize stitch behavior, both experimentally and through multiscale models. In this work, a framework is developed for the geometric modeling of lockstitch architectures and the integration of these stitches with reinforcement material. Homogenization of numerical results will be used to compare with experimental tensile results to evaluate the influence of stitch pattern and placement on the CFRP so that a fundamental understanding of embroidery stitch influenced mechanics can be established. Through these initial mechanical characterizations, it is anticipated that improved topology design and stitch pattern optimization can occur for embroidery stitch-based modification of composite reinforcements.

A comprehensive benchmarking study of a critical aerospace component, the pickle fork derivative, has been performed by fabricating it using various additive-subtractive hybrid manufacturing process routes. The objective was to evaluate and compare the process capability, geometric fidelity, and hybrid manufacturability of emerging deposition and forging technologies for structural applications. The selected processes included wire arc additive manufacturing (WAAM), wire-laser directed energy deposition (DED), laser powder bed fusion (LPBF), powder-blown DED, additive friction stir deposition (AFSD), and agility Forging. Wrought 316 L stainless steel and Al 6061 parts were machined and served as baselines. Each preform was fabricated, scanned using structured light, and subsequently finish-machined following a standardized hybrid processing route to ensure consistent benchmarking. The results reveal distinct process-dependent characteristics in build resolution, surface integrity, and overbuild allowance, which directly influence subsequent machining requirements and achievable dimensional accuracy. Fusion-based processes such as L-PBF and laser wire-DED produced near-net geometries with minimal overbuild, whereas WAAM, AFSD, and agility forging exhibited higher material allowances due to coarser resolution, thermal distortion, and build volume limitations. Despite these variations, all preforms were successfully machined to achieve dimensional conformity with the target CAD geometry, demonstrating the compatibility of additive and hybrid approaches for structural part fabrication. Further, preliminary mechanical performance evaluations were performed for hardness, strength, and fatigue resistance. The study highlights the critical role of process selection and preform accuracy in optimizing hybrid manufacturing workflows, providing key insights for the integration of additive, subtractive, and forging processes in aerospace component production.

A parallelized X-ray diffraction (XRD) approach for rapid oxidation characterization is demonstrated on a multimaterial plate fabricated via laser directed energy deposition (DED). The XRD approach was first validated on DED Inconel 625 (IN625) oxidized in air at 950 degrees C, by comparing the evolution of the oxide fraction at the specimen surface (as determined from fitting XRD patterns) with the mass gain evolution. We then investigated a four-material 2 & times; 2 plate oxidized at 950 degrees C, which added three other alloys: Nimonic 80 A (N80A), Haynes 25 (HA25), and stainless steel (316L). Finally, rapid characterization was demonstrated on a five-material 5 & times; 5 plate oxidized at 950 degrees C, with five replicates of five alloys (IN625, N80A, HA25, 316L, and a 1:1 blend of IN625 + HA25). The oxide fraction determined via XRD is again shown to scale with low-throughput experiments where the mass gain is measured sequentially on individual specimens. The relative fraction of various oxide species (Cr2O3, M3O4, TiO2) was also determined during XRD measurements, showing increasing fractions of M3O4 with oxidation time for all alloys. While all alloys display substantial oxidation resistance due to their high Cr content that forms a protective Cr2O3 scale, N80A and 316L oxidize faster, and also form less-protective oxides, than IN625, HA25, and the mixed alloy, which form mostly Cr-rich oxides with higher oxidation protection.

Compositionally Graded Alloys (CGAs) offer unprecedented design flexibility by enabling spatial variations in composition; tailoring material properties to local loading conditions. This flexibility leads to components that are stronger, lighter, and more cost-effective than traditional monolithic counterparts. The fabrication of CGAs have become increasingly feasible owing to recent advancements in additive manufacturing (AM), particularly in multi-material printing and improved precision in material deposition. However, AM of CGAs requires imposition of manufacturing constraints; in particular limits on the maximum spatial gradation of composition. This paper introduces a topology optimization (TO) based framework for designing optimized CGA components with controlled compositional gradation. In particular, we represent the constrained composition distribution using a band-limited coordinate neural network. By regulating the network's bandwidth, we ensure implicit compliance with gradation limits, eliminating the need for explicit constraints. The proposed approach also benefits from the inherent advantages of TO using coordinate networks, including mesh independence, high-resolution design extraction, and end-to-end differentiability. The effectiveness of our framework is demonstrated through various elastic and thermo-elastic TO examples.

This work presents an incremental open-die forging platform, Agility Forge, to enable flexible high-mix manufacturing. Additionally, the platform can be utilized to characterize material property at high temperature through the differentiability of a GPU-accelerated finite element platform JAX-FEM, which incorporates fully coupled thermomechanical behavior at finite strain and temperature-dependent material parameters. JAX-FORGE, built on JAX-FEM, develops a seamless interface for multi-hit forging operations. The framework is validated by reproducing the Agility Forge robotic forging process and compared with experimental measurements via laser profilometer. JAX-FORGE constitutes a core component of a broader simulation ecosystem for process planning.

Digital Twin-a virtual replica of a physical system enabling real-time monitoring, model updating, prediction, and decision-making-combined with recent advances in machine learning, offers new opportunities for proactive control strategies in autonomous manufacturing. However, achieving real-time decision-making with Digital Twins requires efficient optimization driven by accurate predictions of highly nonlinear manufacturing systems. This paper presents a simultaneous multi-step Model Predictive Control (MPC) framework for realtime decision-making, using a multivariate deep neural network, named Time-Series Dense Encoder (TiDE), as the surrogate model. Unlike conventional MPC models which only provide one-step ahead prediction, TiDE is capable of predicting future states within the prediction horizon in one shot (multi-step), significantly accelerating the MPC. Using Directed Energy Deposition (DED) additive manufacturing as a case study, we demonstrate the effectiveness of the proposed MPC in achieving melt pool temperature tracking to ensure part quality, while reducing porosity defects by regulating laser power to maintain melt pool depth constraints. In this work, we first show that TiDE is capable of accurately predicting melt pool temperature and depth. Second, we demonstrate that the proposed MPC achieves precise temperature tracking while satisfying melt pool depth constraints within a targeted dilution range (10%-30%), reducing potential porosity defects. Compared to Proportional-Integral-Derivative (PID) controller, the MPC results in smoother and less fluctuating laser power profiles with competitive or superior melt pool temperature control performance. This demonstrates the MPC's proactive control capabilities, leveraging time-series prediction and real-time optimization, positioning it as a powerful tool for future Digital Twin applications and real-time process optimization in manufacturing.

Additive manufacturing has enabled the fabrication of functionally graded materials (FGMs), such as compositionally graded alloys (CGAs), offering unprecedented flexibility in structural design. CGAs hold significant potential for thermal-elastic applications, yet existing design methods often overlook temperature-dependent material properties due to the complexity of coupled physics, design-dependent temperature fields, and local constraints. To address these challenges, we propose a topology optimization (TO) framework that concurrently designs geometry and graded material composition while accounting for temperature-dependent material behaviors and nonlinear thermal analysis. Our method employs a radial basis function (RBF)-based interpolation scheme to model material properties as functions of both temperature and material composition. Additionally, we leverage automatic differentiation and adjoint sensitivity analysis for computational efficiency and extensibility to GPU acceleration. Numerical examples demonstrate the effectiveness of our approach, underscoring (1) the critical role of temperature-dependent material properties in thermal-elastic structure optimization and (2) the benefits of continuous material grading in enhancing structural performance.

This study investigates the impact of surface texturing on the durability of slippery liquid-infused porous surface (SLIPS) coatings applied to sheet metal substrates by the double-sided incremental forming (DSIF) process. Toolmarks generated during the DSIF process were leveraged as an efficient method for texturing, enabling both the formation and texturing of surfaces using a single set of universal tools. The effects of texture patterns, spacing, and tool movement on the SLIPS performance were evaluated by comparing samples generated in the presence and absence of tool spinning/rotation. The results indicate that textures with dimple patterns significantly improve coating durability by acting as lubricant reservoirs, reducing oil depletion, and supporting self-healing. In contrast, continuous grooves were less effective due to limited capillary action and increased edge effects. Tool spinning further enhanced the surface topography, producing an undulating texture that minimized contact line pinning and improved the surface hydrophobicity. Low-speed spinning (approximately 10 rpm) facilitated a transition to mixed sliding-rolling friction, resulting in smoother textures and extended coating durability. Combining dimple patterns and controlled spinning provides a synergistic approach for optimizing SLIPS coatings, offering a practical solution for enhancing durability without requiring additional equipment. This study underscores the potential of controlled texturing and tool movement to improve the SLIPS efficacy and broaden its applications in industrial, clinical, and consumer environments.

Triaxial braided laminates have an expanding scope of use across the aerospace and automotive industries due to their quasi-isotropic nature. However, stochastic variations caused by processing or nesting in the micro-and meso-structure can cause anisotropic behavior. Experimentally dependent characterization of these behaviors can be time consuming and expensive, and even more so for novel material systems that are cost-prohibitive to obtain in small batches. Therefore, the development of simulation techniques for predicting braided properties is vital for promoting a more sustainable approach to design. In this study, multiscale techniques are employed to determine the effect of stochastic variations and uncertainty on the orthotropic elastic constants of a triaxial braid. Unlike previous methods relying on calibrated fiber-tow inputs or analytical microstructure solutions, realistic RVEs are generated in this work using novel molecular dynamics fiber packing techniques and thermal-mechanical manipulations. A library of these micro and meso-structure unit cells is assembled and paired to capture the load angle dependent variation from stochastic uncertainty. Results of the simulations conform excellently with experimental validation, and errors in the model's predicted values range between 2 and 10 %. Using the developed method, an analytical homogenization model is calibrated, allowing for targeted inverse parameter design using automatic differentiation in JAX. The sensitivity relationship between material and structural input to orthotropic properties are evaluated and established to allow for inverse RVE parameter design.