Pith. sign in

REVIEW 2 major objections 4 minor 300 references

All reduced-order domain-decomposition methods fit two families

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review · deepseek-v4-flash

2026-08-03 10:32 UTC pith:6PE5W7HS

load-bearing objection Useful engineering-oriented map of ROM+DD, but the binary intrusive/non-intrusive taxonomy is over-claimed and the corpus is not auditable. the 2 major comments →

arxiv 2601.09623 v2 pith:6PE5W7HS submitted 2026-01-14 math.NA cs.NAphysics.flu-dyn

A Structured Review of Reduced Order Modeling for Domain Decomposition Problems: State of the Art and Perspectives

Shenhui Ruan , Andreas G. Class , Gianluigi Rozza This is my paper
classification math.NA cs.NAphysics.flu-dyn MSC 65N5565M99
keywords reduced order modelsdomain decompositionsubdomain couplingintrusive methodsnon-intrusive methodslocal reduced basesSchwarz methodsphysics-informed neural networks
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This review tries to bring order to the many ways researchers combine reduced-order models with domain decomposition, a technique that splits a geometry into subdomains and glues local fast surrogates into a global solution. It claims that, despite dozens of individually adapted techniques, all coupling methods rest on a few concepts and can be sorted into intrusive (projection-based) and non-intrusive (data-driven) families. Within these, it maps monolithic versus iterative coupling and Schwarz-based, interpolation, optimization, and physics-informed neural network approaches. It further argues that training local models on small repeated archetype blocks is the most cost-effective offline strategy, and that a discontinuous-Galerkin-based reduced-basis method is the most promising route for engineering problems. If the review is right, a newcomer can choose a coupling method from the map, and research effort can concentrate on a few recommended families.

Core claim

On its own terms, the review establishes that ROM+DD coupling methods, despite their diversity, rest on a few concepts and sort into two families: intrusive (projection-based) and non-intrusive (data-driven). Intrusive methods split into monolithic schemes (RBEM, RDF, SCRBEM, DGRBEM, partition of unity, optimization-based) and iterative Schwarz-type schemes; non-intrusive methods split into Schwarz-based, interpolation, optimization, and PINN groups. The review further argues that localized training on archetype blocks is the most cost-effective offline strategy, and that DGRBEM — a discontinuous-Galerkin reduced-basis element method whose jump-penalty interface terms glue local bases direct

What carries the argument

The central machinery is the local-ROM assembly pipeline and its two-family classification. Three common preliminaries — domain decomposition, parameterization, and local reduced-basis construction — precede every method. The coupling principle is uniform: minimize discontinuities at interfaces while satisfying the PDEs and boundary conditions. Named workhorses include monolithic RBEM (Lagrange multipliers), RDF (interface finite-element basis), SCRBEM (port/bubble static condensation), DGRBEM (penalized jumps), partition-of-unity weights, optimization-based functionals, and iterative Schwarz, Dirichlet-Neumann, and Robin-Robin schemes. The key organizational identity is the taxonomy itself,

Load-bearing premise

The load-bearing premise is that the cited literature is a representative, faithfully categorized sample of the field; without an auditable search protocol, a skewed corpus would change both the taxonomy's completeness and the 'most promising' ranking.

What would settle it

A systematic literature search with explicit inclusion/exclusion criteria that finds a substantial family of coupling methods missing from the two-category map would falsify the completeness claim. Alternatively, a head-to-head benchmark on a repeating-geometry engineering problem (e.g., a thermal fin or blood-vessel flow) using the same snapshot budget — comparing DGRBEM against RBEM/SCRBEM, Dirichlet-Neumann Schwarz, and a data-driven interpolation method — would settle the 'most promising' claim: if DGRBEM is not the accuracy-per-cost winner, the recommendation fails.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • A researcher entering ROM+DD can choose a coupling method by locating it on the two-family map instead of surveying the full literature.
  • If localized training on archetype blocks is as efficient as claimed, offline costs for large repeating geometries — heat exchangers, nuclear fuel assemblies, vascular networks — can be cut by training on small representative systems and transforming bases to each instance.
  • If DGRBEM is the most promising monolithic method, engineering-oriented ROM implementations will likely standardize on discontinuous-Galerkin assembly with jump penalties for interface continuity.
  • Iterative Schwarz-type couplings are predicted to reach commercial codes before monolithic methods because of their simpler formulation and natural fit with many subdomains and multiphysics.
  • The review implies research attention should move toward coupling many subdomains and toward robust FOM-ROM and ROM-ROM communication schemes.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Editorial extension: the 'most promising' ranking is only as strong as the representativeness of the cited corpus; a different selection of papers could shift the ranking, so the recommendation should be read as a hypothesis about the literature rather than a measured fact.
  • Editorial extension: the review's localized-training premise — small networks represent large-system behavior — is directly testable by training on a small assembly and predicting on progressively larger assemblies to map the accuracy decay with scale.
  • Editorial extension: the taxonomy invites a shared benchmark suite that reports accuracy, offline cost, and online cost for the same repeating-geometry problem across all mapped families; such a benchmark would be the natural next step and would turn the map into a quantitative decision tool.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

2 major / 4 minor

Summary. The paper is a survey of reduced order modeling (ROM) combined with domain decomposition (DD). It proposes a hierarchical taxonomy: common preliminaries (domain decomposition, parameterization, dimensionality reduction), followed by a binary split into intrusive (projection-based) and non-intrusive (data-driven) coupling frameworks. Intrusive methods are further divided into monolithic and iterative schemes, and the review presents the underlying formulations for RBEM, RBHM, RDF, SCRBEM, DG-based methods, partition-of-unity methods, optimization-based methods, and Schwarz-type iterations. Non-intrusive methods are organized into Schwarz-based, interpolation-based, optimization-based, and physics-informed neural network (PINN) groups. The paper concludes that the field can be adequately classified in this way and recommends DGRBEM as the most promising method for engineering problems, with localized training and archetype blocks as the most cost-effective offline strategy.

Significance. If the taxonomy and the engineering recommendations are accepted, the paper provides a useful map of a fragmented literature and could help newcomers choose coupling strategies. Its strengths are the large collection of references, the worked algebraic descriptions for the intrusive families, and the up-to-date treatment of PINN-based domain decomposition, including cPINN, XPINN, DPINN, FBPINN, and related variants. The paper is less useful as a critical guide, however, because the central classification claim and the ‘most promising’ judgments rest on definitional choices and a corpus whose representativeness is not audited.

major comments (2)
  1. [§1 vs. §5.1.3] The central dichotomy is internally inconsistent. In §1, non-intrusive methods are defined as ‘purely data-driven’ methods that ‘adopt a set of sampled data to train a surrogate model.’ In §5.1.3, the paper states that PINNs can solve PDEs ‘without requiring data’ and that the PINN loss includes the PDE residual, yet it classifies PINNs as non-intrusive because they do not ‘explicitly solve the PDE system through algebraic manipulation.’ This changes the defining criterion mid-paper. Under the original definition, PINNs are neither intrusive (no Galerkin projection) nor non-intrusive (not purely data-driven). Because PINN methods are reviewed extensively in §5.5 and are folded into the two-category conclusion in §6, the claimed dichotomy is not exhaustive. A consistent definition—or a third category such as ‘physics-constrained’ or ‘PDE-informed’—is needed, and the summary conclusions sh
  2. [§6 and §1] The paper makes global knowledge claims—‘the available methods can be adequately classified into two categories’ and ‘DGRBEM is the most promising for engineering problems’—but it gives no literature-search protocol, inclusion/exclusion criteria, or coverage statistics. The set of roughly 300 references is therefore not auditable as a representative corpus. The scope note in §1 explicitly excludes clustering-based local ROMs, yet the concluding claim is unqualified. In addition, several of the recommended methods and illustrative examples come from the authors’ own research circle (e.g., refs. [16,74,107,108,135,136,138,177,178,205,217,246,261,291]); without explicit evaluation criteria (offline cost, online speedup, accuracy, implementation maturity, generality), the ‘most promising’ judgment is not transparent. I recommend adding a short methods subsection on corpus construction and ev
minor comments (4)
  1. [Eq. (10)] The sentence following Eq. (10) is incomplete: ‘because w_{m,i}|’ is cut off. Please complete the explanation of why the interface term vanishes for the test space.
  2. [§5 opening] The opening of §5 describes non-intrusive methods as operating ‘without requiring any modification of the underlying governing equations,’ which is a different criterion from the ‘purely data-driven’ definition in §1. This is exactly the ambiguity that leads to the PINN classification problem; the terminology should be harmonized.
  3. [Eq. (7) and §4.1] The notation in Eq. (7) is typeset ambiguously: ‘F(w_i) v_j ∈ V, w_i ∈ W’ should be separated into two clauses. Also, the use of n for both the trial-space dimension and the outer normal is confusing in the same section.
  4. [Headings] Heading capitalization is inconsistent (e.g., ‘parameterization techniques’ vs. ‘Physical Informed Neural Network’). Please standardize.

Circularity Check

0 steps flagged

No circular derivation: the review's taxonomy and recommendations are qualitative summaries of cited work, not predictions forced by fitted inputs or by the authors' own definitions.

full rationale

This manuscript is a narrative literature review and contains no fitted parameters, no numerical predictions, and no first-principles derivation whose output could coincide with an input. Its load-bearing claims are (i) that ROM+DD methods can be organized into intrusive/non-intrusive, monolithic/iterative, and four data-driven families, and (ii) that DGRBEM and localized training are the most promising options. Both are presented as qualitative summaries of the cited literature: equations in §§4-5 are standard formulations taken from the cited methods, not new results derived from the taxonomy. The taxonomy is stipulated, not deduced. The notable tension is that §1 defines non-intrusive methods as 'purely data-driven,' while §5.1.3 admits PINNs 'can still solve PDEs without requiring data' and then reclassifies them as non-intrusive because they 'do not explicitly solve the PDE system through algebraic manipulation.' This is an internal inconsistency in the classification criteria and a legitimate correctness/validity concern, but it is not circularity: the category labels are stipulated, and the paper makes no prediction that is forced by an equation or by a definition. Self-citations (SISSA mathLab/KIT works) are numerous and several illustrative and recommended methods come from that circle, but the load-bearing recommendations (e.g., DGRBEM, oversampling/localized training) are also supported by independent citations such as [5,53,203] and [40,11,129,256], and no uniqueness theorem or ansatz is imported from the authors' own prior work. The absence of a documented literature-search protocol weakens the survey's auditability but does not make its conclusions circular. Hence no circular step is identifiable, and the circularity score is 0.

Axiom & Free-Parameter Ledger

0 free parameters · 3 axioms · 0 invented entities

The paper contributes a taxonomy, not a derivation: there are no free parameters, no new equations fitted to data, and no invented entities. Its dependence is on secondary-source fidelity and on the scope choice (spatial DD only). The three assumptions listed are the ones whose failure would change the conclusions.

axioms (3)
  • domain assumption The surveyed corpus of ~300 references is representative of the ROM+DD literature and is categorized faithfully.
    The taxonomy and the §6 engineering recommendations rest entirely on the completeness and correct reading of the cited papers; the paper gives no search protocol to audit this.
  • domain assumption Spatial domain decomposition, not snapshot clustering, is the right frame for local ROMs.
    §1 explicitly excludes clustering-based classifications (k-means etc., refs [3,4,278,227]); this boundary shapes every conclusion, e.g., the prognosis that generic/archetype decomposition will dominate.
  • domain assumption Localized training presupposes that phenomena in small archetype systems faithfully represent dynamics in the global system.
    §3.3.3 states 'it is assumed that the dynamics that occur in small-scale networks can represent the physical behavior of the original large-scale systems'; the review's endorsement of localized training in §6 inherits this assumption, which the paper itself flags as an assumption.

pith-pipeline@v1.3.0-alltime-deepseek · 50801 in / 10203 out tokens · 107929 ms · 2026-08-03T10:32:13.036283+00:00 · methodology

0 comments
read the original abstract

Reduced Order Models (ROMs) have been regarded as an efficient alternative to conventional high-fidelity Computational Fluid Dynamics (CFD) for accelerating the design and optimization processes in engineering applications. Many industrial geometries feature repeating subdomains or contain sub-regions governed by distinct physical phenomena, making them well-suited to Domain Decomposition (DD) techniques. The integration of ROM and DD is promising to further reduce computational costs by constructing local ROMs and assembling them into global solutions. Due to the complexity and necessity of coupling ROMs, many approaches have been proposed in recent years. This review provides a concise overview of existing methodologies combining ROM and DD. We categorize existing methods into intrusive (projection-based) and non-intrusive (data-driven) frameworks. Various strategies for generating local reduced bases and coupling them across subdomains are illustrated. Particular emphasis is placed on intrusive techniques, including equations, numerical algorithms, and practical implementations. The non-intrusive framework is also discussed, highlighting its general procedures, basic formulations, and underlying principles. Finally, we summarise the state of the literature, identify open challenges, and present perspectives on future implementation from an engineering viewpoint.

Figures

Figures reproduced from arXiv: 2601.09623 by Andreas G. Class, Gianluigi Rozza, Shenhui Ruan.

Figure 1
Figure 1. Figure 1: The procedures and classification for preliminaries of constructing ROMs. Abbreviation: [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Schematic of domain decomposition: (a) overlapping; (b) non-overlapping. The global [PITH_FULL_IMAGE:figures/full_fig_p008_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Domain decomposition into 2 × 2 subdomains. Left: overlapping partition, redrawn based on [79]. Right: non-overlapping partition, redrawn based on [82]. 3.1.2 Conforming and non-conforming meshes The conforming and non-conforming high-resolution meshes along an interface are displayed in [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: (a) Conforming and (b) non-conforming meshes for adjacent subdomains. [PITH_FULL_IMAGE:figures/full_fig_p009_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Examples of domain divisions for problems without repeating geometric parts. (a) K´arm´an [PITH_FULL_IMAGE:figures/full_fig_p009_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Decomposition for fluid-structure interaction problems. (a) Left: fluid; right: solid. Redrawn [PITH_FULL_IMAGE:figures/full_fig_p009_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Several generic parts can be used to assemble entire models. The real blocks are obtained [PITH_FULL_IMAGE:figures/full_fig_p010_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Sampling strategies in the space P = [−1, 1]2 . Grid-based (first row, from left to right): 4 × 4 uniform, 3 × 5 uniform, and 3 × 5 Clenshaw-Curtis. Statistical (second row, from left to right): random (15 points), Latin hypercube (15 points), and Smolyak sparse grid with Clenshaw-Curtis points. Figures redrawn based on [223]. The situation becomes more challenging when the dimension NP > 10. The common sa… view at source ↗
Figure 9
Figure 9. Figure 9: Parameters for a carotid artery bifurcation [ [PITH_FULL_IMAGE:figures/full_fig_p012_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Definition of the parameters of a marine propeller’s blade. The blade is represented by a [PITH_FULL_IMAGE:figures/full_fig_p012_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Generation of geometric samples with control points. Applications for (a) an aircraft wing [PITH_FULL_IMAGE:figures/full_fig_p013_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: Two-dimensional FFD implementation for a bypass structure. FFD deforms a lattice (red) [PITH_FULL_IMAGE:figures/full_fig_p014_12.png] view at source ↗
Figure 13
Figure 13. Figure 13: FFD for local deformation of a cardiovascular structure. e.g., a blood vessel. Figure [PITH_FULL_IMAGE:figures/full_fig_p014_13.png] view at source ↗
Figure 14
Figure 14. Figure 14: Individual domain decomposition and local RBs for a backward-facing step problem. The [PITH_FULL_IMAGE:figures/full_fig_p016_14.png] view at source ↗
Figure 15
Figure 15. Figure 15: Domain decomposition of a thermo-hydro-mechanical system. Many Ω [PITH_FULL_IMAGE:figures/full_fig_p017_15.png] view at source ↗
Figure 16
Figure 16. Figure 16: The model for collecting high-dimensional data and dominant modes for archetypes. Fig [PITH_FULL_IMAGE:figures/full_fig_p018_16.png] view at source ↗
Figure 17
Figure 17. Figure 17: A sketch of the Oversampling strategy for three archetype components, i.e., Ωˆint , Ωˆco and Ωˆed. Three small-scale models are constructed to generate high-fidelity solutions of each generic subdomain. Redrawn based on [256]. 3.3.4 RB for interior, interface, and boundary. We finalize this section by discussing the practical formation of the snapshot’s matrix regarding the interior Ω, interfaces Γ (or th… view at source ↗
Figure 18
Figure 18. Figure 18: Snapshots structures. (a) A domain and three regions: , Ω; interface, Γ; and global [PITH_FULL_IMAGE:figures/full_fig_p019_18.png] view at source ↗
Figure 19
Figure 19. Figure 19: The categorization of intrusive techniques. The abbreviation: [PITH_FULL_IMAGE:figures/full_fig_p019_19.png] view at source ↗
Figure 20
Figure 20. Figure 20: Port and Bubble training. Redraw based on [ [PITH_FULL_IMAGE:figures/full_fig_p025_20.png] view at source ↗
Figure 21
Figure 21. Figure 21: Schematic description of the basis functions employed for Continuous Galerkin (left) and [PITH_FULL_IMAGE:figures/full_fig_p028_21.png] view at source ↗
Figure 22
Figure 22. Figure 22: Sketch of PoU for a one-dimensional (1-D) case showing two subdomains Ω [PITH_FULL_IMAGE:figures/full_fig_p031_22.png] view at source ↗
Figure 23
Figure 23. Figure 23: Sketch and definition of the embedded boundary method. From left to right: the back [PITH_FULL_IMAGE:figures/full_fig_p037_23.png] view at source ↗
Figure 24
Figure 24. Figure 24: The classification of data-driven techniques. The abbreviation: [PITH_FULL_IMAGE:figures/full_fig_p042_24.png] view at source ↗
Figure 25
Figure 25. Figure 25: A sketch of neural network architecture. [PITH_FULL_IMAGE:figures/full_fig_p043_25.png] view at source ↗
Figure 26
Figure 26. Figure 26: Architectures of recurrent neural networks for a sequential dataset. [PITH_FULL_IMAGE:figures/full_fig_p044_26.png] view at source ↗
Figure 27
Figure 27. Figure 27: Example of convolution and pooling layers in the convolutional neural network. Figure [PITH_FULL_IMAGE:figures/full_fig_p044_27.png] view at source ↗
Figure 28
Figure 28. Figure 28: A sketch of an autoencoder. High-fidelity snapshots [PITH_FULL_IMAGE:figures/full_fig_p045_28.png] view at source ↗
Figure 29
Figure 29. Figure 29: Architecture of a PINN [269]. The inputs are coordinates x in the computational domain and time t. The outputs are physical variables u(x, t) over the spatial and temporal domains. The MSE consists of two parts: the PDEs loss and the error with respect to the constraints (initial and boundary conditions). The parameters of the Deep Neural Network, namely w and b, are trained to reduce the combined loss. F… view at source ↗
Figure 30
Figure 30. Figure 30: PINN training data points for a domain to compute loss functions. Figure redrawn based [PITH_FULL_IMAGE:figures/full_fig_p045_30.png] view at source ↗
Figure 31
Figure 31. Figure 31: A subdomain and its neighbours. Redraw based on [ [PITH_FULL_IMAGE:figures/full_fig_p048_31.png] view at source ↗
Figure 32
Figure 32. Figure 32: Global reconstruction using Gappy-POD. The entire domain is decomposed into non [PITH_FULL_IMAGE:figures/full_fig_p051_32.png] view at source ↗
Figure 33
Figure 33. Figure 33: Computational domain and cPINN training data points. Figure redrawn based on [ [PITH_FULL_IMAGE:figures/full_fig_p053_33.png] view at source ↗
Figure 34
Figure 34. Figure 34: Local test functions of VPINN over each subdomain (first row). Results of each subdomain [PITH_FULL_IMAGE:figures/full_fig_p054_34.png] view at source ↗
Figure 35
Figure 35. Figure 35: One dimensional example of the FBPINN solver. (a) Domain decomposition and window [PITH_FULL_IMAGE:figures/full_fig_p056_35.png] view at source ↗
Figure 36
Figure 36. Figure 36: Hierarchy of levels used in the multilevel FBPINN. Figure taken from [ [PITH_FULL_IMAGE:figures/full_fig_p056_36.png] view at source ↗
Figure 37
Figure 37. Figure 37: A sketch of the Distributed Mosaic Flow Predictor. (a) FOM simulations are performed in a slightly larger model, and local snapshots are extracted from the solutions, which is similar to the oversampling technique explained in Section 3.3.3. (b) Mosaic Flow Predictor contains four sub-steps in each iteration. Both atomic and overlapping subdomains overlap with their neighbours, and the solutions are commu… view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

300 extracted references · 30 canonical work pages

  1. [1]

    C. C. Aggarwal et al.Neural networks and deep learning, volume 10. Springer, 2018

    2018

  2. [2]

    Aletti and D

    M. Aletti and D. Lombardi. A reduced-order representation of the Poincar´ e–Steklov operator: an application to coupled multi-physics problems.International Journal for Numerical Methods in Engineering, 111(6):581–600, 2017

    2017

  3. [3]

    Amsallem, J

    D. Amsallem, J. Cortial, and C. Farhat. Towards real-time computational-fluid-dynamics-based aeroelastic computations using a database of reduced-order information.AIAA journal, 48(9): 2029–2037, 2010. 63

    2029

  4. [4]

    Amsallem, M

    D. Amsallem, M. J. Zahr, and C. Farhat. Nonlinear model order reduction based on local reduced-order bases.International Journal for Numerical Methods in Engineering, 92(10):891– 916, 2012

    2012

  5. [5]

    P. F. Antonietti, P. Pacciarini, and A. Quarteroni. A discontinuous Galerkin reduced basis element method for elliptic problems.ESAIM: Mathematical Modelling and Numerical Analysis, 50(2):337–360, 2016

    2016

  6. [6]

    Arcucci, C

    R. Arcucci, C. Q. Casas, D. Xiao, L. Mottet, F. Fang, P. Wu, C. Pain, and Y.-K. Guo. A domain decomposition reduced order model with data assimilation (DD-RODA). InParallel Computing: Technology Trends, pages 189–198. IOS Press, 2020

    2020

  7. [7]

    Arcucci, L

    R. Arcucci, L. Mottet, C. A. Q. Casas, F. Guitton, C. Pain, and Y.-K. Guo. Adaptive domain decomposition for effective data assimilation. InEuro-Par 2019: Parallel Processing Workshops: Euro-Par 2019 International Workshops, G¨ ottingen, Germany, August 26–30, 2019, Revised Selected Papers 25, pages 583–595. Springer, 2020

    2019

  8. [8]

    Babuska and R

    I. Babuska and R. Lipton. Optimal local approximation spaces for generalized finite element methods with application to multiscale problems.Multiscale Modeling & Simulation, 9(1):373– 406, 2011

    2011

  9. [9]

    Babuˇ ska and J

    I. Babuˇ ska and J. M. Melenk. The partition of unity method.International journal for numerical methods in engineering, 40(4):727–758, 1997

    1997

  10. [10]

    Babuˇ ska, G

    I. Babuˇ ska, G. Caloz, and J. E. Osborn. Special finite element methods for a class of second order elliptic problems with rough coefficients.SIAM Journal on Numerical Analysis, 31(4): 945–981, 1994

    1994

  11. [11]

    Babuˇ ska, R

    I. Babuˇ ska, R. Lipton, P. Sinz, and M. Stuebner. Multiscale-spectral GFEM and optimal over- sampling.Computer Methods in Applied Mechanics and Engineering, 364:112960, 2020

    2020

  12. [12]

    Bai and L

    Z. Bai and L. Peng. Non-intrusive nonlinear model reduction via machine learning approxima- tions to low-dimensional operators.Advanced Modeling and Simulation in Engineering Sciences, 8(1):28, 2021

    2021

  13. [13]

    Baiges, R

    J. Baiges, R. Codina, and S. Idelsohn. A domain decomposition strategy for reduced order mod- els. application to the incompressible Navier–Stokes equations.Computer Methods in Applied Mechanics and Engineering, 267:23–42, 2013

    2013

  14. [14]

    Balajewicz and C

    M. Balajewicz and C. Farhat. Reduction of nonlinear embedded boundary models for problems with evolving interfaces. 274:489–504. ISSN 00219991. doi: 10.1016/j.jcp.2014.06.038. URL https://linkinghub.elsevier.com/retrieve/pii/S0021999114004458

    doi:10.1016/j.jcp.2014.06.038 2014

  15. [15]

    Ballarin

    F. Ballarin. Reduced-order models for patient-specific haemodynamics of coronary artery bypass grafts. 2015

    2015

  16. [16]

    Ballarin and G

    F. Ballarin and G. Rozza. POD-Galerkin monolithic reduced order models for parametrized fluid-structure interaction problems.International Journal for Numerical Methods in Fluids, 82 (12):1010–1034, Dec. 2016. ISSN 0271-2091, 1097-0363. doi: 10.1002/fld.4252. Publisher: Wiley

    work page doi:10.1002/fld.4252 2016

  17. [17]

    Ballarin, A

    F. Ballarin, A. Manzoni, G. Rozza, and S. Salsa. Shape Optimization by Free-Form Deformation: Existence Results and Numerical Solution for Stokes Flows.Journal of Scientific Computing, 60(3):537–563, Sept. 2014. ISSN 0885-7474, 1573-7691. doi: 10.1007/s10915-013-9807-8

    work page doi:10.1007/s10915-013-9807-8 2014

  18. [18]

    Ballarin, E

    F. Ballarin, E. Faggiano, S. Ippolito, A. Manzoni, A. Quarteroni, G. Rozza, and R. Scrofani. Fast simulations of patient-specific haemodynamics of coronary artery bypass grafts based on a POD-Galerkin method and a vascular shape parametrization.Journal of Computational Physics, 315:609–628, June 2016. ISSN 00219991. doi: 10.1016/j.jcp.2016.03.065. 64

    work page doi:10.1016/j.jcp.2016.03.065 2016

  19. [19]

    Ballarin, A

    F. Ballarin, A. D’Amario, S. Perotto, and G. Rozza. A POD-selective inverse distance weighting method for fast parametrized shape morphing.International Journal for Numerical Methods in Engineering, 117(8):860–884, Feb. 2019. ISSN 0029-5981, 1097-0207. doi: 10.1002/nme.5982

    work page doi:10.1002/nme.5982 2019

  20. [20]

    Ballarin, S

    F. Ballarin, S. Lee, and S.-Y. Yi. Projection-based reduced order modeling of an iterative scheme for linear thermo-poroelasticity.Results in Applied Mathematics, 21:100430, Feb. 2024. ISSN 25900374. doi: 10.1016/j.rinam.2023.100430

    arXiv 2024

  21. [21]

    Barnett, I

    J. Barnett, I. Tezaur, and A. Mota. The Schwarz alternating method for the seamless coupling of nonlinear reduced order models and full order models.arXiv preprint arXiv:2210.12551, 2022

    Pith/arXiv arXiv 2022

  22. [22]

    Barzegar Gerdroodbary and S

    M. Barzegar Gerdroodbary and S. Salavatidezfouli. A predictive surrogate model based on linear and nonlinear solution manifold reduction in cardiovascular FSI: A comparative study. Computers in Biology and Medicine, 189:109959, May 2025. ISSN 0010-4825. doi: 10.1016/j. compbiomed.2025.109959. Publisher: Elsevier BV

    arXiv 2025

  23. [23]

    Batta and A

    A. Batta and A. Class. CFD analysis of pressure drop across grid spacers in rod bundles compared to correlations and heavy liquid metal experimental data.Nuclear Engineering and Design, 312:121–127, 2017

    2017

  24. [24]

    Batta and A

    A. Batta and A. G. Class. CFD validation of forced and natural convection for the open phase of IAEA benchmark CRP-I31038.Arabian Journal for Science and Engineering, pages 1–7, 2024

    2024

  25. [25]

    F. B. Belgacem. The mortar finite element method with Lagrange multipliers.Numerische Mathematik, 84:173–197, 1999

    1999

  26. [26]

    Benaceur and A

    A. Benaceur and A. Patera. Port-reduced reduced-basis component method for steady state Navier–Stokes and passive scalar equations. 2022

    2022

  27. [27]

    Benner, S

    P. Benner, S. Gugercin, and K. Willcox. A survey of projection-based model reduction methods for parametric dynamical systems.SIAM review, 57(4):483–531, 2015

    2015

  28. [28]

    Benner, W

    P. Benner, W. Schilders, S. Grivet-Talocia, A. Quarteroni, G. Rozza, and L. Miguel Silveira. Model order reduction: volume 3 applications. De Gruyter, 2020

    2020

  29. [29]

    Benner, W

    P. Benner, W. Schilders, S. Grivet-Talocia, A. Quarteroni, G. Rozza, and L. Miguel Silveira. Model Order Reduction: Volume 2: Snapshot-Based Methods and Algorithms. De Gruyter, 2020

    2020

  30. [30]

    T. L. Bergman.Fundamentals of heat and mass transfer. John Wiley & Sons, 2011

    2011

  31. [31]

    Bernardi

    C. Bernardi. A new nonconforming approach to domain decomposition: the mortar element method.Technical Report, Universite Pierre at Marie Curie, 1990

    1990

  32. [32]

    Bonaccorso.Machine Learning Algorithms: Popular algorithms for data science and machine learning

    G. Bonaccorso.Machine Learning Algorithms: Popular algorithms for data science and machine learning. Packt Publishing Ltd, 2018

    2018

  33. [33]

    B. N. Bond and L. Daniel. A piecewise-linear moment-matching approach to parameterized model-order reduction for highly nonlinear systems.IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 26(12):2116–2129, 2007

    2007

  34. [34]

    S. P. Bordas, A. Menk, and S. Natarajan.Partition of unity methods. John Wiley & Sons, 2023

    2023

  35. [35]

    N. W. Bressloff. Parametric geometry exploration of the human carotid artery bifurcation. Journal of Biomechanics, 40(11):2483–2491, 2007

    2007

  36. [36]

    S. L. Brunton and J. N. Kutz.Data-driven science and engineering: Machine learning, dynamical systems, and control. Cambridge University Press, 2024

    2024

  37. [37]

    Buffoni, H

    M. Buffoni, H. Telib, and A. Iollo. Iterative methods for model reduction by domain decompo- sition.Computers & Fluids, 38(6):1160–1167, 2009. 65

    2009

  38. [38]

    B¨ uhler and C

    L. B¨ uhler and C. Mistrangelo. Geometric optimization of electrically coupled liquid metal man- ifolds for WCLL blankets.IEEE Transactions on Plasma Science, 2024

    2024

  39. [39]

    M. D. Buhmann. Radial basis functions.Acta numerica, 9:1–38, 2000

    2000

  40. [40]

    A. Buhr, L. Iapichino, M. Ohlberger, S. Rave, F. Schindler, and K. Smetana. Localized model reduction for parameterized problems. InHandbook on Model Order Reduction. Walter De Gruyter, 2020

    2020

  41. [41]

    Bui-Thanh, K

    T. Bui-Thanh, K. Willcox, and O. Ghattas. Model Reduction for Large-Scale Systems with High- Dimensional Parametric Input Space.SIAM Journal on Scientific Computing, 30(6):3270–3288, Jan. 2008. ISSN 1064-8275, 1095-7197. doi: 10.1137/070694855

    doi:10.1137/070694855 2008

  42. [42]

    Canuto, M

    C. Canuto, M. Y. Hussaini, A. Quarteroni, and T. A. Zang.Spectral methods, volume 285. Springer, 2006

    2006

  43. [43]

    Canuto, M

    C. Canuto, M. Y. Hussaini, A. Quarteroni, and T. A. Zang.Spectral methods: evolution to complex geometries and applications to fluid dynamics. Springer Science & Business Media, 2007

    2007

  44. [44]

    Carlberg, M

    K. Carlberg, M. Barone, and H. Antil. Galerkin v. least-squares Petrov–Galerkin projection in nonlinear model reduction.Journal of Computational Physics, 330:693–734, 2017

    2017

  45. [45]

    Chaturantabut and D

    S. Chaturantabut and D. C. Sorensen. Nonlinear model reduction via discrete empirical inter- polation.SIAM Journal on Scientific Computing, 32(5):2737–2764, 2010

    2010

  46. [46]

    L. Chen, S. Jin, and G. Bu. Experimental investigation of a novel multi-tank thermal energy storage system for solar-powered air conditioning.Applied Thermal Engineering, 123:953–962, 2017

    2017

  47. [47]

    S. Chen, Z. Ding, Q. Li, and S. J. Wright. A reduced order Schwarz method for nonlinear mul- tiscale elliptic equations based on two-layer neural networks.arXiv preprint arXiv:2111.02280, 2021

    Pith/arXiv arXiv 2021

  48. [48]

    W. Chen, J. S. Hesthaven, B. Junqiang, Y. Qiu, Z. Yang, and Y. Tihao. Greedy nonintrusive reduced order model for fluid dynamics.AIAA Journal, 56(12):4927–4943, 2018

    2018

  49. [49]

    Y. Chen, J. S. Hesthaven, and Y. Maday. A seamless reduced basis element method for 2D Maxwell’s problem: an introduction. InSpectral and High Order Methods for Partial Differ- ential Equations: Selected papers from the ICOSAHOM’09 conference, June 22-26, Trondheim, Norway, pages 141–152. Springer, 2011

    2011

  50. [50]

    Y. Chen, Y. Lin, X. Sun, C. Yuan, and Z. Gao. Tensor decomposition-based neural operator with dynamic mode decomposition for parameterized time-dependent problems.Journal of Computational Physics, 533:113996, 2025

    2025

  51. [51]

    Chung, Y

    E. Chung, Y. Efendiev, and T. Y. Hou. Adaptive multiscale model reduction with generalized multiscale finite element methods.Journal of Computational Physics, 320:69–95, 2016

    2016

  52. [52]

    E. T. Chung, Y. Efendiev, and W. T. Leung. Generalized multiscale finite element methods for wave propagation in heterogeneous media.Multiscale Modeling & Simulation, 12(4):1691–1721, 2014

    2014

  53. [53]

    S. W. Chung, Y. Choi, P. Roy, T. Moore, T. Roy, T. Y. Lin, D. T. Nguyen, C. Hahn, E. B. Duoss, and S. E. Baker. Train small, model big: Scalable physics simulators via reduced order modeling and domain decomposition.Computer Methods in Applied Mechanics and Engineering, 427:117041, 2024. 66

    2024

  54. [54]

    Cicci, S

    L. Cicci, S. Fresca, S. Pagani, A. Manzoni, A. Quarteroni, et al. Projection-based reduced order models for parameterized nonlinear time-dependent problems arising in cardiac mechanics. Mathematics in Engineering, 5(2):1–38, 2023

    2023

  55. [55]

    Cinquegrana, R

    D. Cinquegrana, R. Donelli, and A. Viviani. A hybrid method based on POD and domain decomposition to compute the 2-D aerodynamics flow field.AIMETA, Bologna, Italy, pages 1–10, 2011

    2011

  56. [56]

    Class, M

    A. Class, M. Viellieber, N. Moussiopoulos, and F. Barmpas. Two way coupled micro/meso scale method for wind farms.PAMM, 14(1):585–586, 2014

    2014

  57. [57]

    A. G. Class, M. O. Viellieber, and S. R. Himmel. Coarse Grid CFD for underresolved simulation. InAPS Division of Fluid Dynamics Meeting Abstracts, volume 63 ofAPS Meeting Abstracts, nov 2010

    2010

  58. [58]

    C. W. Clenshaw and A. R. Curtis. A method for numerical integration on an automatic com- puter.Numerische Mathematik, 2:197–205, 1960

    1960

  59. [59]

    Cockburn

    B. Cockburn. Discontinuous Galerkin methods for computational fluid dynamics.Encyclopedia of computational mechanics, 2004

    2004

  60. [60]

    Cockburn, G

    B. Cockburn, G. E. Karniadakis, and C.-W. Shu.Discontinuous Galerkin methods: theory, computation and applications, volume 11. Springer Science & Business Media, 2012

    2012

  61. [61]

    Colciago, S

    C. Colciago, S. Deparis, and A. Quarteroni. Comparisons between reduced order models and full 3D models for fluid-structure interaction problems in haemodynamics.Journal of Computational and Applied Mathematics, 265:120–138, Aug. 2014. ISSN 0377-0427. doi: 10.1016/j.cam.2013. 09.049. Publisher: Elsevier BV

    work page doi:10.1016/j.cam.2013 2014

  62. [62]

    P. G. Constantine.Active subspaces: Emerging ideas for dimension reduction in parameter studies. SIAM, 2015

    2015

  63. [63]

    Corigliano, M

    A. Corigliano, M. Dossi, and S. Mariani. Domain decomposition and model order reduction methods applied to the simulation of multi-physics problems in MEMS.Computers & Structures, 122:113–127, 2013

    2013

  64. [64]

    Corigliano, M

    A. Corigliano, M. Dossi, S. Mariani, et al. Combined domain decomposition and model order reduction methods for the solution of coupled and non-linear problems. InProceedings of 11th. World Congress on Computational Mechanics (WCCM XI); 5th. European Congress on Com- putational Mechanics (ECCM V); 6th. European Congress on Computational Fluid Dynamics (EC...

    2014

  65. [65]

    Corigliano, M

    A. Corigliano, M. Dossi, and S. Mariani. Model order reduction and domain decomposition strategies for the solution of the dynamic elastic–plastic structural problem.Computer Methods in Applied Mechanics and Engineering, 290:127–155, 2015

    2015

  66. [66]

    Coscia, A

    D. Coscia, A. Ivagnes, N. Demo, and G. Rozza. Physics-informed neural networks for advanced modeling.Journal of Open Source Software, 8(87):5352, 2023

    2023

  67. [67]

    Courtessole, H.-J

    C. Courtessole, H.-J. Brinkmann, L. B¨ uhler, and J. Roth. Experimental investigation of MHD flows in a WCLL TBM mock-up.Fusion Engineering and Design, 202:114306, 2024

    2024

  68. [68]

    T. L. D. Croft.Proper generalised decompositions: theory and applications. PhD thesis, Cardiff University, 2015

    2015

  69. [69]

    Cuomo, V

    S. Cuomo, V. S. Di Cola, F. Giampaolo, G. Rozza, M. Raissi, and F. Piccialli. Scientific machine learning through physics–informed neural networks: Where we are and what’s next.Journal of Scientific Computing, 92(3):88, 2022. 67

    2022

  70. [70]

    D’Amario.A Reduced-Order Inverse Distance Weighting Technique for the Efficient Mesh- Motion of Deformable Interfaces and Moving Shapes in Computational Problems

    A. D’Amario.A Reduced-Order Inverse Distance Weighting Technique for the Efficient Mesh- Motion of Deformable Interfaces and Moving Shapes in Computational Problems. PhD thesis, Politecnico di Milano, 2014

    2014

  71. [71]

    de Castro, P

    A. de Castro, P. Kuberry, I. Tezaur, and P. Bochev. A lagrange multiplier partitioned scheme for coupling reduced order models. Technical report, Sandia National Laboratories and Stanford University, 2022

    2022

  72. [72]

    A. G. de Castro, P. A. Kuberry, I. K. Tezaur, and P. B. Bochev. A partitioned method for reduced ordermodel-finite element model (ROM-FEM) and ROM-ROM couplings with separate reduced bases for interior and interface variables. Technical report, Sandia National Laboratories and Stanford University, 2023

    2023

  73. [73]

    N. Demo, M. Tezzele, and G. Rozza. EZyRB: Easy Reduced Basis method.The Journal of Open Source Software, 3(24):661, 2018. doi: https://doi.org/10.21105/joss.00661

    doi:10.21105/joss.00661 2018

  74. [74]

    N. Demo, M. Tezzele, and G. Rozza. A non-intrusive approach for the reconstruction of POD modal coefficients through active subspaces.Comptes Rendus. M´ ecanique, 347(11):873–881, Nov. 2019. ISSN 1873-7234. doi: 10.1016/j.crme.2019.11.012

    doi:10.1016/j.crme.2019.11.012 2019

  75. [75]

    N. Demo, M. Tezzele, A. Mola, and G. Rozza. Hull Shape Design Optimization with Parameter Space and Model Reductions, and Self-Learning Mesh Morphing.Journal of Marine Science and Engineering, 9(2):185, Feb. 2021. ISSN 2077-1312. doi: 10.3390/jmse9020185

    doi:10.3390/jmse9020185 2021

  76. [76]

    Deparis, D

    S. Deparis, D. Forti, P. Gervasio, and A. Quarteroni. INTERNODES: an accurate interpolation- based method for coupling the Galerkin solutions of PDEs on subdomains featuring non- conforming interfaces.Computers & Fluids, 141:22–41, Dec. 2016. ISSN 00457930. doi: 10.1016/j.compfluid.2016.03.033

    work page doi:10.1016/j.compfluid.2016.03.033 2016

  77. [77]

    Diab and M

    W. Diab and M. Al-Kobaisi. Temporal neural operator for modeling time-dependent physical phenomena.arXiv preprint arXiv:2504.20249, 2025

    Pith/arXiv arXiv 2025

  78. [78]

    A. N. Diaz, Y. Choi, and M. Heinkenschloss. Nonlinear-manifold reduced order models with domain decomposition.arXiv preprint arXiv:2312.00713, 2023

    Pith/arXiv arXiv 2023

  79. [79]

    A. N. Diaz, Y. Choi, and M. Heinkenschloss. A fast and accurate domain decomposition nonlinear manifold reduced order model.Computer Methods in Applied Mechanics and Engineering, 425: 116943, 2024

    2024

  80. [80]

    Z. Ding, K. Wang, J. Li, and X. Cong. Flow field and oscillatory shear stress in a tuning- fork-shaped model of the average human carotid bifurcation.Journal of Biomechanics, 34(12): 1555–1562, 2001

    2001

Showing first 80 references.