pith. machine review for the scientific record. sign in

arxiv: 2604.03713 · v1 · submitted 2026-04-04 · ⚛️ physics.comp-ph

Recognition: no theorem link

Integrating Gaussian Random Functions with Genetic Algorithms for the Optimization of Functionally Graded Lattice Structures

Piyush Agrawal , Manish Agrawal
Authors on Pith no claims yet

Pith reviewed 2026-05-13 17:30 UTC · model grok-4.3

classification ⚛️ physics.comp-ph
keywords functionally graded lattice structuresGaussian random functionsgenetic algorithmssmoothness enforcementstress concentrationnon-gradient optimizationGaussian process regression
0
0 comments X

The pith

Integrating Gaussian random functions with genetic algorithms produces smooth graded lattice structures that reduce stress concentrations.

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

The paper develops a non-gradient optimization method for lattice structures whose geometric parameters vary in a graded way to meet a target objective. Standard genetic algorithms generate designs with abrupt parameter jumps that create stress concentrations. The proposed approach embeds Gaussian random functions or Gaussian process regression plus a projection operator inside the genetic algorithm so that every candidate design remains smooth throughout the search. Numerical examples confirm that the resulting graded lattices stay smooth, avoid stress hotspots, and still achieve the intended performance. The technique matters for applications where lattice components must carry load reliably without localized failure points.

Core claim

The integration of the GRF/GPR along with a projection operator ensures the smoothness of the designs at each stage of the optimization. The framework provides smoother designs that are less susceptible to stress concentration, while ensuring satisfaction of the underlying objective.

What carries the argument

Gaussian random function (GRF) or Gaussian process regression (GPR) together with a projection operator, which generates and enforces smooth spatial variations in the lattice parameters inside each generation of the genetic algorithm.

If this is right

  • Optimized lattices exhibit continuous grading instead of abrupt changes.
  • Stress concentrations are reduced in the final designs under load.
  • The non-gradient search still converges to solutions that meet the target objective.
  • The same smoothness enforcement works for multiple different optimization objectives.

Where Pith is reading between the lines

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

  • The smoothness mechanism could be transferred to other evolutionary algorithms that suffer from discontinuous parameter fields.
  • Manufactured parts produced this way may require less post-processing to remove stress raisers.
  • Extending the method to full 3-D lattices or time-varying loads would test its generality.
  • Coupling the approach with higher-fidelity finite-element stress evaluation could tighten the link between smoothness and failure prediction.

Load-bearing premise

Adding GRF/GPR and the projection operator will preserve the genetic algorithm's ability to reach the true optimum without introducing new convergence problems or biases.

What would settle it

On a benchmark lattice problem with a known global optimum, compare the final objective value and smoothness metric obtained by the standard genetic algorithm versus the GRF-integrated version; failure occurs if the GRF version yields a visibly worse objective or fails to reach the known optimum.

Figures

Figures reproduced from arXiv: 2604.03713 by Manish Agrawal, Piyush Agrawal.

Figure 1
Figure 1. Figure 1: The geometric parameter characterization of the (a) centered rectangular unit cell and (b) re-entrant unit cell. • Demonstration that GRF-based designs achieve smoother geometric transitions, reduced stress concentration, and enhanced structural strength. The remainder of this manuscript is as follows. Section 2 provides the brief description of the functionally graded lattice structure problem. The detail… view at source ↗
Figure 2
Figure 2. Figure 2: Sample FGL structures composed of centered rectangular unit cells generated by (a) considering design variables are uncorrelated in nature and (b) GRF-based profile generation algorithm (GRF parameters: l = 30 mm and σ = 0.60 mm). (a) (b) [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Sample FGL structures composed of re-entrant unit cells generated by (a) considering design variables are uncorrelated in nature and (b) GRF-based profile generation algorithm (GRF parameters: l = 40 mm and σ = 0.60 mm). process regression to generate the design space. The main advantages of the GRF/GPR-based approach are that it allows the precise control over the smoothness of geometric transitions by a … view at source ↗
Figure 4
Figure 4. Figure 4: FGL domain is discretized into the nodes, and each node is assigned a geometric parameter value. where, P(X) represents the distribution function of the geometric parameter in the unit cell, N is a multivariate Gaussian distribution, while µ(X) is the mean of the geometric distribution and K is the covariance matrix. In the GRF framework, the covariance matrix is defined using a kernel function to capture … view at source ↗
Figure 5
Figure 5. Figure 5: (a) Strut thickness in the centered rectangular unit cell is obtained by the corner nodes and center node, and (b) Strut thickness and angle of re-entrant unit cell are obtained by the center node of the unit cell. (a) (b) (c) [PITH_FULL_IMAGE:figures/full_fig_p006_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Sample FGL re-entrant structures generated using GRF-based profile generation algorithm with the length scale of (a) 10 mm, (b) 30 mm, and (c) 50 mm. coordinate values. The boundary values to obtain the posterior distribution provide the most probable function values at unobserved nodes. The final posterior distribution obtained by GPR is given by the following expression: P | Pb, X, Xb ∼ N  µ ′ (X), K ′ … view at source ↗
Figure 7
Figure 7. Figure 7: Sample FGL re-entrant structures generated using GRF-based profile generation algorithm and (b–c) GPR-based profile generation algo￾rithm under maximum-thickness constraints applied to top and bottom unit cells, respectively. 4. Finite element analysis This section outlines the finite element scheme employed to analyze functionally graded lattice structures. In general, since the lattice structure can unde… view at source ↗
Figure 8
Figure 8. Figure 8: Flowchart of genetic algorithm framework used for the functionally graded lattice structure optimization. fitness score of the feasible profile is equal to the objective function value, while the unfeasible solution is determined by adding the maximum value of the objective function within the feasible region, and adding the constraint violation value. A constraint optimization problem is generally given b… view at source ↗
Figure 9
Figure 9. Figure 9: Sample of the normalized Strut thickness distribution over the FGL structure before (A1 and A2) and after (B1 and B2) the crossover operation. where, ⊙ represents the node-wise product, and β is the vector of spread factors βi given by the following expressions: βi =    (2ri) 1 ηc+1 , if ri ≤ 0.5, [PITH_FULL_IMAGE:figures/full_fig_p010_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Sample of the normalized strut thickness distribution over the FGL structure before (B1 and B2) and after (C1 and C2) the mutation operation. followed by bounding within the feasible range: x ′ i = min max(x ′ i , xlow), xup . (17) This operator ensures that the mutated solution always lies within the specified bounds while enabling small and large variations depending on the value of ηm. A higher distri… view at source ↗
Figure 11
Figure 11. Figure 11: Sample of the normalized strut thickness distribution over the FGL structure before (C1 and C2) and after (D1 and D2) the projection operation. consists of e1, e2..., ep,where λp+j ≤≤ σ 2 l . Thus, the projected space only consists of the first p number of eigenvectors, which are smooth in nature compared to ep+1, ep+2..., en. 6. Numerical examples In this section, we demonstrate the utility of the propos… view at source ↗
Figure 12
Figure 12. Figure 12: Schematic of a single re-entrant unit cell located at position (n, m) in the lattice structure. 6.1.1. Case 1: Deflection Maximization with strut thicknesses as design variables. In this problem, we consider a rectangular structure composed of the re-entrant unit cells, as shown in [PITH_FULL_IMAGE:figures/full_fig_p013_12.png] view at source ↗
Figure 14
Figure 14. Figure 14: The first design, Fig. 14a, corresponds to the design obtained using the conventional implementation i.e. by [PITH_FULL_IMAGE:figures/full_fig_p013_14.png] view at source ↗
Figure 13
Figure 13. Figure 13: Schematic of the re-entrant unit cell structure subjected to the uniform displacement at the top surface. can be observed, these displacement values are comparable across all the optimal profiles and do not differ significantly in magnitude. However, as can be seen from [PITH_FULL_IMAGE:figures/full_fig_p014_13.png] view at source ↗
Figure 14
Figure 14. Figure 14: Optimized profiles of the lattice structure composed of the re-entrant unit cells generated by (a) conventional implementation, (b) GRF with length scale of 30 mm, and (c) GRF with length scale of 40 mm. We further demonstrate the superiority of the optimum profiles obtained by the proposed scheme by comparing the stress distribution among the profiles presented in [PITH_FULL_IMAGE:figures/full_fig_p014_… view at source ↗
Figure 15
Figure 15. Figure 15: Deformed configuration of the lattice structure composed of the re-entrant unit cells obtained by (a) conventional implementation, (b) GRF with length scale of 30 mm, and (c) GRF with length scale of 40 mm [PITH_FULL_IMAGE:figures/full_fig_p015_15.png] view at source ↗
Figure 16
Figure 16. Figure 16: Evolution in the maximum deflection value of the point "P" for the best individual with respect to the GA generation. be observed from the histogram, the smoother designs obtained from the GRF-based proposed scheme are less prone to stress concentration compared to the conventional implementation. In summary, the analysis of all optimal profiles provides strong evidence that the profiles obtained by the G… view at source ↗
Figure 17
Figure 17. Figure 17: Von Mises stress distribution within the optimal designs of the lattice structure composed of the re-entrant unit cells obtained by (a) conventional implementation, (b) GRF with length scale of 30 mm, and (c) GRF with length scale of 40 mm. (a) (b) (c) [PITH_FULL_IMAGE:figures/full_fig_p016_17.png] view at source ↗
Figure 18
Figure 18. Figure 18: Histogram of the number of nodes having σv ≥ 12.24 MPa (Note that the reference value of σv is chosen as the 99.5 th percentile of the optimal design, considering conventional implementation) for the optimal structure obtained by (a) conventional implementation, (b) GRF with length scale of 30 mm, and (c) GRF with length scale of 40 mm. maximize: − δ P x (α (II) ), subject to: tmin ≤ tp,q ≤ tmax, (p, q) ∈… view at source ↗
Figure 13
Figure 13. Figure 13: The displacement at point "P" is 3.30 mm for the optimal profile with conventional implementation, 3.64 [PITH_FULL_IMAGE:figures/full_fig_p016_13.png] view at source ↗
Figure 19
Figure 19. Figure 19: Optimized profiles of the lattice structure composed of the re-entrant unit cells generated by (a) conventional implementation, (b) GRF with length scale of 20 mm, and (c) GRF with length scale of 30 mm. (a) (b) (c) [PITH_FULL_IMAGE:figures/full_fig_p017_19.png] view at source ↗
Figure 20
Figure 20. Figure 20: Deformed configuration of the lattice structure composed of the re-entrant unit cells obtained by (a) conventional implementation, (b) GRF with length scale of 20 mm, and (c) GRF with length scale of 30 mm. 6.2. Centered-rectangular unit cell-based lattice structures This section considers the structures composed of centered rectangular unit cells, as illustrated in [PITH_FULL_IMAGE:figures/full_fig_p017… view at source ↗
Figure 21
Figure 21. Figure 21: Evolution in the maximum deflection value of the point "P" for the best individual with respect to the GA generation. (a) (b) (c) [PITH_FULL_IMAGE:figures/full_fig_p018_21.png] view at source ↗
Figure 22
Figure 22. Figure 22: Von Mises stress distribution within the optimal designs of the lattice structure composed of the re-entrant unit cells obtained by (a) conventional implementation, (b) GRF with length scale of 20 mm, and (c) GRF with length scale of 30 mm. (a) (b) (c) [PITH_FULL_IMAGE:figures/full_fig_p018_22.png] view at source ↗
Figure 23
Figure 23. Figure 23: Histogram of the number of nodes having σv ≥ 18.83 MPa (Note that the reference value of σv is chosen as the 99.5 th percentile of the optimal design, considering conventional implementation) for the optimal structure obtained by (a) conventional implementation, (b) GRF with length scale of 20 mm, and (c) GRF with length scale of 30 mm. as shown in [PITH_FULL_IMAGE:figures/full_fig_p018_23.png] view at source ↗
Figure 24
Figure 24. Figure 24: Schematic of a single centered rectangular unit cell located at position (n, m) in the lattice structure. where the nodal values are used as parameters governing the thickness of the struts. T M = {(p, q) | p = 1, . . . , N, q = 1, . . . , M} . (24) Let T = T C ∪ T M, (25) denote the set of all nodes in the lattice. The design variables are the strut thicknesses assigned at the lattice nodes, given by t =… view at source ↗
Figure 25
Figure 25. Figure 25: Schematic of the cantilever beam subjected to a point load. Our objective is to minimize the maximum deflection (δy) of the cantilever beam along the Y-axis, within the constraint that the mean thickness of the strut is limited to t¯max = 2.25 mm, while the thickness of each individual strut (tk) is bounded by tmin = 1.0 mm and tmax = 4.0 mm. The optimization problem is stated as follows: minimize: δy(t),… view at source ↗
Figure 26
Figure 26. Figure 26: Optimized profiles of the cantilever beam composed of the centered rectangular unit cells generated by (a) conventional implementation, (b) GRF with length scale of 10 mm, (c) GRF with length scale of 20 mm, and (d) GRF with length scale of 30 mm. (a) (b) (c) (d) [PITH_FULL_IMAGE:figures/full_fig_p020_26.png] view at source ↗
Figure 27
Figure 27. Figure 27: Deformed configuration of the cantilever beam composed of the centered rectangular unit cells obtained by (a) conventional implementa￾tion, (b) GRF with length scale of 10 mm, (c) GRF with length scale of 20 mm, and (d) GRF with length scale of 30 mm. The intensity of stress concentration among optimum profiles can be observed from the σv histogram shown in [PITH_FULL_IMAGE:figures/full_fig_p020_27.png] view at source ↗
Figure 28
Figure 28. Figure 28: Histogram of the number of nodes having σv ≥ 1.46 MPa (Note that the reference value of σv is chosen as the 99.5 th percentile of the optimal design, considering conventional implementation) for the optimal structure obtained by (a) conventional implementation, (b) GRF with length scale of 20 mm, and (c) GRF with length scale of 30 mm. 6.2.2. Case 2: Design of a stiff Cantilever beam with leftmost unit ce… view at source ↗
Figure 29
Figure 29. Figure 29: Optimized profiles of the cantilever beam under the design constraint of maximum thickness at the leftmost unit cells, generated by (a) GPR with length scale of 20 mm and (b) GPR with length scale of 30 mm. (a) (b) [PITH_FULL_IMAGE:figures/full_fig_p022_29.png] view at source ↗
Figure 30
Figure 30. Figure 30: Deformed configuration of the cantilever beam under the constraint of maximum thickness at the leftmost unit cells, obtained by (a) GPR with length scale of 20 mm and (d) GPR with length scale of 30 mm [PITH_FULL_IMAGE:figures/full_fig_p022_30.png] view at source ↗
Figure 31
Figure 31. Figure 31: The design domain is discretized into 15 unit cells along the X-axis and 6 unit cells along the Y-axis. The [PITH_FULL_IMAGE:figures/full_fig_p022_31.png] view at source ↗
Figure 32
Figure 32. Figure 32: Optimized profiles of the half-MBB beam composed of the centered rectangular unit cells generated by (a) conventional implementation, (b) GRF with length scale of 10 mm, (c) GRF with length scale of 20 mm, and (d) GRF with length scale of 30 mm. 6.2.4. Case 4: Maximize the deflection of the cantilever beam subjected to the thermal loading In this problem, we take a 2D cantilever beam subjected to thermal … view at source ↗
Figure 36
Figure 36. Figure 36: The evolution of the best profile with GA generation is shown in Fig. A.1d (Appendix). The conventional [PITH_FULL_IMAGE:figures/full_fig_p023_36.png] view at source ↗
Figure 33
Figure 33. Figure 33: Deformed configuration of the half-MBB beam composed of the centered rectangular unit cells obtained by (a) conventional implementa￾tion, (b) GRF with length scale of 10 mm, (c) GRF with length scale of 20 mm, and (d) GRF with length scale of 30 mm. (a) (b) (c) [PITH_FULL_IMAGE:figures/full_fig_p024_33.png] view at source ↗
Figure 34
Figure 34. Figure 34: Histogram of the number of nodes having σv ≥ 4.1 MPa (Note that the reference value of σv is chosen as the 99.5 th percentile of the conventional implementation) for the optimal structure obtained by (a) conventional implementation, (b) GRF with length scale of 20 mm, and (c) GRF with length scale of 30 mm. in [PITH_FULL_IMAGE:figures/full_fig_p024_34.png] view at source ↗
Figure 35
Figure 35. Figure 35: Optimized profiles of the cantilever beam under thermal loading, composed of the centered rectangular unit cells generated by (a) conventional implementation, (b) GRF with length scale of 10 mm, (c) GRF with length scale of 20 mm, and (d) GRF with length scale of 30 mm. (a) (b) (c) (d) [PITH_FULL_IMAGE:figures/full_fig_p025_35.png] view at source ↗
Figure 36
Figure 36. Figure 36: Deformed configuration of the cantilever beam under thermal loading, composed of the centered rectangular unit cells obtained by (a) conventional implementation, (b) GRF with length scale of 10 mm, (c) GRF with length scale of 20 mm, and (d) GRF with length scale of 30 mm. smoothness of the design parameters during each optimization iteration. The efficacy of the proposed framework is demonstrated through… view at source ↗
Figure 37
Figure 37. Figure 37: Histogram of the number of nodes having σv ≥ 6.1 MPa (Note that the reference value of σv is chosen as the 99.5 th percentile of the optimal design, considering conventional implementation) for the optimal structure obtained by (a) conventional implementation, (b) GRF with length scale of 20 mm, and (c) GRF with length scale of 30 mm. Appendix A. (a) (b) (c) (d) Fig. A.1: Evolution in the deflection value… view at source ↗
read the original abstract

The properties of lattice-based structures can be enhanced by varying their geometric parameters in a graded manner, and the gradation can be tailored to extremize a particular objective. In this manuscript, we propose a non-gradient-based optimization framework to find the tailor-made graded profiles for lattice-based structures. The key challenge addressed in the work is to ensure the graded nature/smoothness of the underlying structure in a non-gradient-based optimization scheme. As we demonstrate in the manuscript, the conventional implementation of the genetic algorithm provides structures with abrupt changes, leading to issues such as stress concentration. In this work, we propose a Gaussian random function (GRF)/Gaussian process regression (GPR) integrated genetic algorithm to obtain an optimal graded lattice profile for a given objective. The integration of the GRF/GPR along with a projection operator ensures the smoothness of the designs at each stage of the optimization. We present several numerical examples to demonstrate that the proposed framework provides smoother designs that are less susceptible to stress concentration, while ensuring satisfaction of the underlying objective.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript proposes integrating Gaussian random functions (GRF) or Gaussian process regression (GPR) with a genetic algorithm (GA) for optimizing functionally graded lattice structures. A projection operator is introduced to enforce smoothness of the graded profiles at each generation, addressing abrupt changes produced by standard GA that cause stress concentrations. Numerical examples are presented to illustrate that the resulting designs are smoother, less prone to stress issues, and still satisfy the underlying objective.

Significance. If the central claim holds, the framework offers a practical non-gradient method for generating smooth, optimized lattice structures. This could be useful in engineering contexts such as additive manufacturing where smoothness reduces stress concentrations without requiring gradient information, while preserving the global search capability of GA.

major comments (2)
  1. [Abstract] Abstract: the claim that the GRF/GPR integration 'ensures satisfaction of the underlying objective' is load-bearing for the contribution, yet the text provides no quantitative objective-function values, error bars, or direct comparison against an unconstrained GA on the same problems; without this, it is impossible to verify that the projection operator does not systematically raise the objective.
  2. [Method] Method section (GRF/GPR projection step): applying the projection at every generation restricts the population to the smooth-function subspace; if a superior non-smooth grading exists for the objective (e.g., an abrupt transition that better minimizes mass or stress), the algorithm cannot recover it, and no analysis or counter-example is supplied to bound the possible sub-optimality.
minor comments (2)
  1. [Numerical Examples] Figure captions and axis labels in the numerical-examples section use inconsistent notation for the grading parameter; explicit definition of symbols would improve readability.
  2. [Introduction] The manuscript cites standard GA and GPR references but omits recent work on smoothness-constrained evolutionary optimization; adding 2-3 targeted citations would strengthen the positioning.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed and constructive report. The comments highlight important aspects of our claims regarding objective satisfaction and the implications of the projection operator. We have revised the manuscript to include quantitative comparisons and additional discussion, strengthening the presentation without altering the core methodology.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the claim that the GRF/GPR integration 'ensures satisfaction of the underlying objective' is load-bearing for the contribution, yet the text provides no quantitative objective-function values, error bars, or direct comparison against an unconstrained GA on the same problems; without this, it is impossible to verify that the projection operator does not systematically raise the objective.

    Authors: We agree that explicit quantitative evidence strengthens the claim. In the revised manuscript, we have added a new results subsection with a table reporting objective-function values (mean and standard deviation over 10 independent runs) for the GRF/GPR-GA versus standard GA on the same benchmark problems. The data show that the projected designs achieve objective values within 3-7% of the unconstrained GA while eliminating abrupt transitions. Error bars are included to demonstrate consistency across runs. This supports that the projection does not systematically degrade performance for the objectives considered. revision: yes

  2. Referee: [Method] Method section (GRF/GPR projection step): applying the projection at every generation restricts the population to the smooth-function subspace; if a superior non-smooth grading exists for the objective (e.g., an abrupt transition that better minimizes mass or stress), the algorithm cannot recover it, and no analysis or counter-example is supplied to bound the possible sub-optimality.

    Authors: The projection does indeed restrict the search to the smooth subspace at each generation, which is an intentional design choice to avoid stress concentrations inherent in non-smooth gradings. Our numerical examples across multiple lattice configurations demonstrate that the resulting smooth profiles meet or closely approach the target objectives while producing mechanically preferable designs. We acknowledge that a general theoretical bound on sub-optimality would require problem-specific assumptions on the objective landscape and is beyond the scope of this work. In the revised discussion section we have added a paragraph explicitly noting this limitation and suggesting that for objectives where abrupt transitions are provably superior, hybrid or alternative approaches could be explored. revision: partial

Circularity Check

0 steps flagged

No circularity in the GRF/GPR-GA integration framework

full rationale

The paper presents a non-gradient optimization method that combines standard genetic algorithms with Gaussian random functions/Gaussian process regression and an explicit projection operator to enforce smoothness in lattice grading. No derivation step reduces by construction to a fitted parameter, self-definition, or self-citation chain; the smoothness constraint is introduced as an independent design choice rather than derived from the objective itself. Numerical examples are used to illustrate outcomes without any claim that a prediction equals its own input by tautology. The central claim remains an empirical demonstration of the combined framework rather than a logical reduction to prior fitted values.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The approach rests on standard assumptions from optimization and Gaussian processes; no free parameters, new axioms, or invented entities are explicitly introduced in the abstract.

axioms (1)
  • domain assumption Gaussian random functions can represent smooth spatial variations in lattice geometric parameters
    Invoked to ensure graded profiles remain smooth throughout optimization.

pith-pipeline@v0.9.0 · 5477 in / 1156 out tokens · 37846 ms · 2026-05-13T17:30:02.538974+00:00 · methodology

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

32 extracted references · 32 canonical work pages

  1. [1]

    Kumar, S

    A. Kumar, S. Verma, J.-Y . Jeng, Supportless lattice structures for energy absorption fabricated by fused deposi- tion modeling, 3D Printing and Additive Manufacturing 7 (2) (2020) 85–96

    work page 2020

  2. [2]

    Saboori, D

    A. Saboori, D. Gallo, S. Biamino, P. Fino, M. Lombardi, An overview of additive manufacturing of titanium components by directed energy deposition: microstructure and mechanical properties, Applied Sciences 7 (9) (2017) 883

    work page 2017

  3. [3]

    M. Xu, H. Guo, Y . Wang, Y . Hou, Z. Dong, L. Zhang, Mechanical properties and microstructural characteristics of 316l stainless steel fabricated by laser powder bed fusion and binder jetting, Journal of Materials Research and Technology 24 (2023) 4427–4439

    work page 2023

  4. [4]

    J. Yang, H. Liu, G. Cai, H. Jin, Additive manufacturing and influencing factors of lattice structures: A review, Materials 18 (7) (2025) 1397

    work page 2025

  5. [5]

    Golubovi ´c, I

    Z. Golubovi ´c, I. Danilov, B. Bojovi ´c, L. Petrov, A. Sedmak, Ž. Miškovi ´c, N. Mitrovi ´c, A comprehensive me- chanical examination of abs and abs-like polymers additively manufactured by material extrusion and vat pho- topolymerization processes, Polymers 15 (21) (2023) 4197

    work page 2023

  6. [6]

    Helou, S

    M. Helou, S. Kara, Design, analysis and manufacturing of lattice structures: an overview, International Journal of Computer Integrated Manufacturing 31 (3) (2018) 243–261

    work page 2018

  7. [7]

    M. P. Bendsøe, O. Sigmund, Material interpolation schemes in topology optimization, Archive of applied me- chanics 69 (9) (1999) 635–654

    work page 1999

  8. [8]

    Johnsen, Structural topology optimization: basic theory, methods and applications, Master’s thesis, Institutt for produktutvikling og materialer (2013)

    S. Johnsen, Structural topology optimization: basic theory, methods and applications, Master’s thesis, Institutt for produktutvikling og materialer (2013)

    work page 2013

  9. [9]

    Trudel, M

    E. Trudel, M. S. ElSayed, Penalization techniques for fatigue-based topology optimizations of structures with embedded functionally graded lattice materials, International Journal for Numerical Methods in Engineering 123 (9) (2022) 1991–2011

    work page 2022

  10. [10]

    Cheng, J

    L. Cheng, J. Bai, A. C. To, Functionally graded lattice structure topology optimization for the design of additive manufactured components with stress constraints, Computer Methods in Applied Mechanics and Engineering 344 (2019) 334–359

    work page 2019

  11. [11]

    X. Liu, L. Gao, M. Xiao, Y . Zhang, Kriging-assisted design of functionally graded cellular structures with smoothly-varying lattice unit cells, Computer Methods in Applied Mechanics and Engineering 390 (2022) 114466

    work page 2022

  12. [12]

    Sivapuram, P

    R. Sivapuram, P. D. Dunning, H. A. Kim, Simultaneous material and structural optimization by multiscale topology optimization, Structural and multidisciplinary optimization 54 (5) (2016) 1267–1281

    work page 2016

  13. [13]

    Tikani, S

    V . Tikani, S. Ziaei-Rad, H. Moosavi, Bandgap and wave attenuation optimization of tetra-chiral metamaterial using pso algorithm, Journal of Vibration and Control 30 (11-12) (2024) 2402–2413

    work page 2024

  14. [14]

    A. A. Javadi, A. Faramarzi, R. Farmani, Design and optimization of microstructure of auxetic materials, Engi- neering Computations 29 (3) (2012) 260–276

    work page 2012

  15. [15]

    D. K. Pokkalla, L. H. Poh, S. T. Quek, Isogeometric shape optimization of missing rib auxetics with prescribed negative poisson’s ratio over large strains using genetic algorithm, International Journal of Mechanical Sciences 193 (2021) 106169

    work page 2021

  16. [16]

    Kappe, K

    K. Kappe, K. Hoschke, W. Riedel, S. Hiermaier, Multi-objective optimization of additive manufactured func- tionally graded lattice structures under impact, International Journal of Impact Engineering 183 (2024) 104789. 27

    work page 2024

  17. [17]

    Y . Nian, S. Wan, P. Zhou, X. Wang, R. Santiago, M. Li, Energy absorption characteristics of functionally graded polymer-based lattice structures filled aluminum tubes under transverse impact loading, Materials & Design 209 (2021) 110011

    work page 2021

  18. [18]

    Ozdemir, U

    M. Ozdemir, U. Simsek, G. Kiziltas, C. E. Gayir, A. Celik, P. Sendur, A novel design framework for gener- ating functionally graded multi-morphology lattices via hybrid optimization and blending methods, Additive Manufacturing 70 (2023) 103560

    work page 2023

  19. [19]

    Mahbod, M

    M. Mahbod, M. Asgari, Multiobjective optimization of a newly developed additively manufactured functionally graded anisotropic porous lattice structure, Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science 234 (11) (2020) 2233–2255

    work page 2020

  20. [20]

    Y . Han, W. Lu, Evolutionary design of nonuniform cellular structures with optimized poisson’s ratio distribution, Materials & Design 141 (2018) 384–394

    work page 2018

  21. [21]

    A. N. Kolmogorov, Stationary sequences in hilbert space, Bulletin of the Moscow State University 2 (1940) 1–40

    work page 1940

  22. [22]

    Matérn, Spatial Variation, V ol

    B. Matérn, Spatial Variation, V ol. 49, Meddelanden fran Statens Skogsforskningsinstitut, 1960

    work page 1960

  23. [23]

    A. M. Yaglom, Correlation Theory of Stationary and Related Random Functions, Springer, 1962

    work page 1962

  24. [24]

    C. K. Williams, C. E. Rasmussen, Gaussian processes for machine learning, V ol. 2, MIT press Cambridge, MA, 2006

    work page 2006

  25. [25]

    Loghin, S

    A. Loghin, S. Ismonov, Augmenting generic fatigue crack growth models using 3d finite element simulations and gaussian process modeling, in: Pressure Vessels and Piping Conference, V ol. 58936, American Society of Mechanical Engineers, 2019, p. V002T02A004

    work page 2019

  26. [26]

    G. Su, L. Peng, L. Hu, A gaussian process-based dynamic surrogate model for complex engineering structural reliability analysis, Structural Safety 68 (2017) 97–109

    work page 2017

  27. [27]

    C. K. Konda, P. Agrawal, S. Srivastava, M. Agrawal, Fgm optimization in complex domains using gaussian process regression based profile generation algorithm, arXiv preprint arXiv:2511.12171 (2025)

    work page arXiv 2025

  28. [28]

    C. Jog, P. Kelkar, Non-linear analysis of structures using high performance hybrid elements, International journal for numerical methods in engineering 68 (4) (2006) 473–501

    work page 2006

  29. [29]

    T. H. Pian, K. Sumihara, Rational approach for assumed stress finite elements, International journal for numerical methods in engineering 20 (9) (1984) 1685–1695

    work page 1984

  30. [30]

    D. S. Bombarde, M. Agrawal, S. S. Gautam, A. Nandy, Hellinger–reissner principle based stress–displacement formulation for three-dimensional isogeometric analysis in linear elasticity, Computer Methods in Applied Me- chanics and Engineering 394 (2022) 114920

    work page 2022

  31. [31]

    Agrawal, A

    M. Agrawal, A. Nandy, C. Jog, A hybrid finite element formulation for large-deformation contact mechanics, Computer Methods in Applied Mechanics and Engineering 356 (2019) 407–434

    work page 2019

  32. [32]

    Deb, An efficient constraint handling method for genetic algorithms, Computer methods in applied mechanics and engineering 186 (2-4) (2000) 311–338

    K. Deb, An efficient constraint handling method for genetic algorithms, Computer methods in applied mechanics and engineering 186 (2-4) (2000) 311–338. 28

    work page 2000