Expensive models and experiments
PIML can reduce repeated simulation or experimental cost while retaining the governing knowledge used in Additive Manufacturing.
Engineering Research Community
Additive Manufacturing & Physics-Informed Machine Learning
From energy deposition and melt pools to qualified parts and closed-loop production
Additive Manufacturing (AM) creates components layer by layer from a digital model. Important process families include laser powder bed fusion (LPBF), directed energy deposition (DED), electron-beam processes, binder jetting, material extrusion, vat photopolymerisation and wire-arc additive manufacturing. AM is simultaneously a manufacturing, thermal, fluid, materials and control problem. Final quality depends on interactions among energy input, feedstock, melt-pool behaviour, solidification, microstructure,…
Physics-Informed Machine Learning (PIML) is especially relevant because AM experiments and high-fidelity simulations are expensive, while substantial physical knowledge is already available.
This page presents ten focused research areas, degree-level project pathways, selected publications and direct support through the PIMLS biweekly members meeting.
This Additive Manufacturing guide covers Physics-Informed Neural Networks (PINNs), physics-guided machine learning, scientific machine learning, neural operators, hybrid models and engineering digital twins. Explore the research and project pathways below, then join the Physics-Informed Machine Learning Society to connect with the international PIMLS community.
The central ideaProcess physics + thermal history + material evidence + machine learning
10focused research areas
3academic project pathways
6selected publications
Biweeklymember research meeting
Why this combination matters
Why Additive Manufacturing Needs Physics-Informed Learning
Use available scientific knowledge to make limited data more useful, transparent and testable.
Incomplete engineering models
Learn uncertain parameters, closures or discrepancies around an inspectable mechanistic foundation.
Transfer across conditions
Test whether structured models generalize across geometries, materials, assets, operating regimes or sites.
Trustworthy evidence
Use physical residuals, independent measurements, uncertainty and conventional engineering baselines before deployment.
Ten focused directions
Major Additive Manufacturing PIML Research Areas
Each card connects a meaningful Additive Manufacturing question with suitable scientific knowledge, modelling choices and evidence needed to test it.
01
Laser Powder Bed Fusion
LPBF involves rapid, local heating and cooling, repeated layer deposition and complex interactions among powder, laser and shielding gas. PIML can support temperature prediction, melt-pool geometry, keyhole/lack-of-fusion boundaries, layerwise monitoring and rapid process mapping. Promising inputs include…
Model and evidenceGoverning equations, calibrated measurements and held-out operating conditions
02
Directed Energy Deposition
DED adds powder or wire into a melt pool. Important variables include power, travel speed, powder/wire feed, standoff distance, shielding gas and substrate geometry. Physics-guided features can capture specific energy, mass deposition, dilution and thermal accumulation. PIML is valuable for bead geometry,…
Model and evidenceMechanistic and data-only baselines, uncertainty and independent validation
03
Wire-Arc Additive Manufacturing
WAAM combines arc heat input, droplet/metal transfer, fluid flow, layer geometry and thermomechanical distortion. Hybrid models can learn uncertain arc efficiency, bead-shape corrections or inter-layer effects around analytical or finite-element models.
Model and evidenceGeometry, material or system parameters, sensor data and physical residuals
04
Polymer and Composite AM
Material extrusion and polymer AM involve heat transfer, viscous flow, cooling, crystallisation, bonding and shrinkage. Fibre-filled or continuous-fibre materials introduce anisotropy and path-dependent properties. PIML opportunities include inter-bead bonding, temperature history, warpage, voids and…
Model and evidenceGoverning equations, calibrated measurements and held-out operating conditions
05
Vat Photopolymerisation
Stereolithography and related processes are governed by light transport, cure kinetics, resin chemistry, heat generation and shrinkage. Knowledge-informed learning can combine exposure models and reaction kinetics with measurements to estimate cure depth, dimensional error and final properties.
Model and evidenceMechanistic and data-only baselines, uncertainty and independent validation
06
Binder Jetting and Sintering
Binder saturation, droplet spreading, powder packing, drying, debinding and sintering affect dimensional accuracy and density. A useful PIML study can connect process physics across printing and post-processing instead of treating sintering shrinkage as an unrelated regression target.
Model and evidenceGeometry, material or system parameters, sensor data and physical residuals
07
Thermal-Field Reconstruction
Full-field temperature is rarely measured directly. Infrared cameras, pyrometers and photodiodes provide incomplete observations with emissivity and line-of-sight limitations. PINNs and data assimilation can reconstruct hidden thermal states while satisfying heat-transfer constraints.
Model and evidenceGoverning equations, calibrated measurements and held-out operating conditions
08
Melt-Pool Geometry and Dynamics
Melt-pool size determines bonding and influences solidification. Direct CFD is expensive, particularly across a process window. Reduced physics plus learned discrepancy or neural operators may provide fast surrogates for optimisation and monitoring.
Model and evidenceMechanistic and data-only baselines, uncertainty and independent validation
09
Defect Formation
Lack of fusion, keyhole porosity, gas porosity, cracking and balling arise through different mechanisms. A strong model should distinguish mechanisms and use appropriate physical indicators. Classification accuracy alone is insufficient when class imbalance or machine-specific correlations dominate.
Model and evidenceGeometry, material or system parameters, sensor data and physical residuals
10
Microstructure Evolution
Thermal gradients and solidification rates influence grain morphology, texture and phase selection. PIML can combine thermal simulations, CALPHAD-informed features, phase-field data and microscopy. The research question should identify whether physics informs the features, architecture, loss or training data.
Model and evidenceGoverning equations, calibrated measurements and held-out operating conditions
Choose the right research depth
Projects for Every Academic Stage
Start with a scope that matches your time, mathematical background, experimental access and expected research contribution.
Project pathway 1
B.E./B.Tech
Learn the foundations with a bounded, measurable system.
- Introductory Laser Powder Bed Fusion model with a clear baseline and validation dataset
- Introductory Directed Energy Deposition model with a clear baseline and validation dataset
- Introductory Wire-Arc Additive Manufacturing model with a clear baseline and validation dataset
- Introductory Polymer and Composite AM model with a clear baseline and validation dataset
- Introductory Vat Photopolymerisation model with a clear baseline and validation dataset
- Introductory Binder Jetting and Sintering model with a clear baseline and validation dataset
Expected outcome
A reproducible implementation, clear baselines, a manageable dataset and physically meaningful validation.
Project pathway 2
M.Tech
Combine an engineering model, substantial data and rigorous comparison.
- Hybrid or inverse Wire-Arc Additive Manufacturing study across unseen operating conditions
- Hybrid or inverse Polymer and Composite AM study across unseen operating conditions
- Hybrid or inverse Vat Photopolymerisation study across unseen operating conditions
- Hybrid or inverse Binder Jetting and Sintering study across unseen operating conditions
- Hybrid or inverse Thermal-Field Reconstruction study across unseen operating conditions
- Hybrid or inverse Melt-Pool Geometry and Dynamics study across unseen operating conditions
Expected outcome
A thesis-quality study with held-out regimes, mechanistic and data-only baselines, ablation and uncertainty.
Project pathway 3
Ph.D.
Address a publishable methodological, multiscale or deployment research gap.
- Transferable, uncertainty-aware research framework for Vat Photopolymerisation
- Transferable, uncertainty-aware research framework for Binder Jetting and Sintering
- Transferable, uncertainty-aware research framework for Thermal-Field Reconstruction
- Transferable, uncertainty-aware research framework for Melt-Pool Geometry and Dynamics
- Transferable, uncertainty-aware research framework for Defect Formation
- Transferable, uncertainty-aware research framework for Microstructure Evolution
Expected outcome
New methodology or validated engineering insight, multi-regime evidence, reproducible software and journal publications.
From idea to evidence
A Strong PIML Project Workflow
01
Define
Choose one Additive Manufacturing question and a measurable engineering output.
02
Model
State the governing relationships, constraints or validated domain knowledge you will retain.
03
Compare
Build mechanistic and data-only baselines before the hybrid model.
04
Validate
Hold out experiments, conditions, assets, sites or regimes at the deployment level.
05
Publish
Report uncertainty, ablation, limitations, data lineage and reproducible code.
Read before you modelDiscuss Your Literature
Selected Publications and Why They Matter
Use this focused reading list to understand the general PIML framework, direct Additive Manufacturing evidence and suitable hybrid modelling methods.
Literature review advice
Do not list papers only. Compare the engineering question, incorporated knowledge, data, split strategy, baselines, uncertainty and evidence level.
This source is included in the Additive Manufacturing literature guide because it demonstrates or reviews a relevant physics-informed, hybrid, inverse, surrogate or scientific-machine-learning approach. Read the methods, data split, baselines and validation evidence—not only the reported accuracy.
How to use this paper: Use this paper to refine the research question, identify a defensible physical prior and compare evidence requirements for Additive Manufacturing.
Read publication or record This source is included in the Additive Manufacturing literature guide because it demonstrates or reviews a relevant physics-informed, hybrid, inverse, surrogate or scientific-machine-learning approach. Read the methods, data split, baselines and validation evidence—not only the reported accuracy.
How to use this paper: Use this paper to refine the research question, identify a defensible physical prior and compare evidence requirements for Additive Manufacturing.
Read publication or record This source is included in the Additive Manufacturing literature guide because it demonstrates or reviews a relevant physics-informed, hybrid, inverse, surrogate or scientific-machine-learning approach. Read the methods, data split, baselines and validation evidence—not only the reported accuracy.
How to use this paper: Use this paper to refine the research question, identify a defensible physical prior and compare evidence requirements for Additive Manufacturing.
Read publication or record This source is included in the Additive Manufacturing literature guide because it demonstrates or reviews a relevant physics-informed, hybrid, inverse, surrogate or scientific-machine-learning approach. Read the methods, data split, baselines and validation evidence—not only the reported accuracy.
How to use this paper: Use this paper to refine the research question, identify a defensible physical prior and compare evidence requirements for Additive Manufacturing.
Read publication or record This source is included in the Additive Manufacturing literature guide because it demonstrates or reviews a relevant physics-informed, hybrid, inverse, surrogate or scientific-machine-learning approach. Read the methods, data split, baselines and validation evidence—not only the reported accuracy.
How to use this paper: Use this paper to refine the research question, identify a defensible physical prior and compare evidence requirements for Additive Manufacturing.
Read publication or record This source is included in the Additive Manufacturing literature guide because it demonstrates or reviews a relevant physics-informed, hybrid, inverse, surrogate or scientific-machine-learning approach. Read the methods, data split, baselines and validation evidence—not only the reported accuracy.
How to use this paper: Use this paper to refine the research question, identify a defensible physical prior and compare evidence requirements for Additive Manufacturing.
Read publication or record Build an interdisciplinary team
Where Additive Manufacturing Can Collaborate
Mechanical Engineering
Thermal, fluid, structural and manufacturing models.
Materials & Metallurgy
Solidification, microstructure, defects and properties.
Computer Science
Operators, vision, optimization and digital twins.
Control & Instrumentation
In-situ sensing, state estimation and closed-loop control.
Before you begin
Frequently Asked Research Questions
These answers help students avoid common scope, terminology and validation mistakes.
Still have a question?
Use the biweekly meeting form for research guidance.
Request accessNo. Additive Manufacturing projects may use physics-guided features, hybrid residual models, differentiable simulators, neural operators, constrained architectures or data assimilation. State exactly what knowledge is incorporated.
Choose one engineering question, a measurable output and a defensible mechanistic baseline. Add learning only where data can identify an uncertainty or discrepancy.
A meaningful question, justified prior knowledge, deployment-level holdouts, strong baselines, ablation, uncertainty, reproducibility and honest limitations.
Simulation can broaden coverage, but simulation-only evidence cannot establish real-system accuracy. Use calibrated experiments, field measurements or trusted independent references appropriate to the claim.
Submit the biweekly members meeting form to discuss your project level, branch, data, model, validation plan and possible collaborators.
Take the next step
Bring your Additive Manufacturing research idea to PIMLS
Join the biweekly members meeting for project guidance, collaboration and publication planning—or contact the Society directly.
Meeting participation is requested through the Google form. Complete it carefully so the Society can understand your research interest.