Pith. sign in

REVIEW 4 major objections 3 minor 80 references

Thermodynamics and Legendre Duality in Optimal Networks

T0 review · 4 major / 3 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read A Legendre-dual pair of generalized potentials governs optimality in nonlinear transport networks, unifying entropy-production and dissipation principles as special cases.

desk verdict A clean synthesis of generalized transport potentials with a useful thermodynamic analogy, but Section 7's apparent-resistance exponent is algebraically wrong for general power laws and needs fixing. read the letter →

arxiv 2506.15727 v1 pith:BBGPN4SX submitted 2025-06-09 cond-mat.stat-mech nlin.AOphysics.app-ph

Amilcare Porporato , Shashank Kumar Anand , Salvatore Calabrese , Luca Ridolfi , Lamberto Rondoni This is my paper
classification cond-mat.stat-mechnlin.AOphysics.app-ph MSC 80A0582C3594C1590B1049Q22
keywords generalizedtransportpotentialsLegendredualitynonequilibriumthermodynamicsoptimalnetworksentropyproductionprinciplesMurray'slawbranchedstabilityandphasetransitions
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

Nonlinear transport networks—electrical circuits, pipe flows, heat conductors—are usually analyzed with constitutive laws that link force and current. The paper claims that the steady operating point of such a network is selected by extremizing a Lagrangian built from a pair of generalized transport potentials, the content $\Phi(J)$ and the co-content $\Psi(X)$, which are Legendre transforms of each other at the operating point. This puts nonequilibrium transport on the same formal footing as equilibrium thermodynamics: the first derivatives of the potentials give the constitutive laws, and the second derivatives give resistances and conductances. The well-known principles of maximum or minimum entropy production and minimum dissipation emerge only as special cases for power-law resistances (with isothermal conditions for entropy production), which explains why they fail for general nonlinear laws. The same potentials also determine the stability of multiple operating points and, when resistances themselves are optimized under a maintenance cost, lead to branched tree networks in the regime of unstable apparent transport laws.

What carries the argument

The machinery is the pair of generalized transport potentials: the content $\Phi(J)=\int_0^J X(J')\,dJ'$ in the flux representation and the co-content $\Psi(X)=\int_0^X J(X')\,dX'$ in the force representation. Their role is to convert the constitutive law into the first-order condition of a variational problem: maximizing $L_J = XJ - \Phi(J)$ over $J$ (or $L_X = XJ - \Psi(X)$ over $X$) returns $X=R(J)J$ at the operating point. The Legendre duality $\Psi = XJ - \Phi$ at that point gives the thermodynamic structure—equations of state $\partial\Psi/\partial X = J$, $\partial\Phi/\partial J = X$ and transport properties $C = \partial^2\Psi/\partial X^2$, $R = \partial^2\Phi/\partial J^2$—and the second variation of the Lagrangians provides the stability criterion. In networks, summing branch potentials and adding current conservation yields reduced Lagrangians and gradient-flow evolution equations toward the operating point.

What would settle it

Take a flux-driven branch whose constitutive law has two steady solutions (the plasma boundary layer of Sec. 6.2 is one). Drive the flux through the critical value $J_c$ and record which branch the system actually settles on. The paper predicts the selected branch is the one maximizing $L_X = XJ - \Psi(X)$; if the system instead follows the maximum-dissipation branch, the variational selection criterion fails.

Watch

Extended reading notes

Core claim

The central claim is that for a branch with force $X$ and current $J$, the operating point $X=R(J)J$ is exactly the stationary point of the Lagrangian $L_J(J)=XJ-\Phi(J)$, where $\Phi(J)=\int_0^J X(J')dJ'$, and dually of $L_X(X)=XJ-\Psi(X)$, where $\Psi(X)=\int_0^X J(X')dX'$. At that operating point the two potentials satisfy the Legendre identity $\Psi(X)=XJ-\Phi(J)$, so the generalized power input $XJ$ acts as the generator connecting the flux and force representations. The paper extends this to networks by summing branch potentials and imposing current conservation, yielding reduced Lagrangians whose maxima select the network operating point, and gradient-flow equations toward it. For power-law resistances the potentials become proportional to generalized dissipation, which is why entropy-production and dissipation extremizations appear to work in those cases; for generic nonlinear laws they do not. The second variation of the same Lagrangians decides which of multiple steady solutions is stable, so changes of stability across a critical force or flux are interpreted as dynamic phase transitions, and a cost-based optimization of resistances reproduces Murray's law and generates branched optimal transport when the apparent resistance exponent is negative.

Load-bearing premise

The paper's load-bearing premise is that virtual changes in current or force that do not satisfy the transport law are physically meaningful, so that the extremum of the resulting Lagrangian over these off-law configurations actually selects the real operating point.

Editorial extensions

If this is right

  • For power-law resistances, maximizing the reduced Lagrangian is equivalent to minimizing generalized dissipation, and under isothermal conditions to the maximum- or minimum-entropy-production principles depending on the imposed constraints; these are therefore special cases of the GTP formalism, not universal selection laws.
  • For generic nonlinear constitutive laws, extremizing dissipation or entropy production generally selects the wrong operating point, so the full Lagrangians $L_J$ and $L_X$ are needed.
  • When a constitutive law has multiple steady solutions, stability is set by whether the corresponding Lagrangian extremum is a maximum or a minimum; crossing a critical force or flux switches stability, which the paper interprets as a dynamic phase transition.
  • Optimizing resistances under a power-law maintenance cost yields an apparent transport law with exponent $\check{\alpha}=(\beta-\alpha)/(1+\beta)$; for $\check{\alpha}<1$ the optimal networks are trees with pruned loops, while for $\check{\alpha}>1$ they are balanced looped networks.
  • The nonequilibrium availability $-L$ measures how far a configuration is from the operating point and provides gradient-flow evolution equations toward it.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • One testable extension is to compare GTP-based and entropy-based selection on a memristive or other strongly nonlinear element, where $\Phi$ is not proportional to dissipation; the two criteria disagree, so the experiment would separate the formalisms.
  • The availability analogy suggests defining a thermodynamic length in current/force space that quantifies the distance between two nonequilibrium configurations; the paper mentions but does not develop this metric, and it could be tested through fluctuation measurements.
  • The ensemble-equivalence comment implies that fluctuation statistics around an operating point may depend on whether currents or forces are held fixed, even though the operating point itself is the same; small-network experiments could look for such a difference.
  • Under the Murray-cost analysis, biological and geophysical networks with sub-additive maintenance costs should generically prune loops and approach tree-like configurations; the paper draws the connection to observed vasculature and river networks but does not confront it with data.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 3 minor

Summary. The paper proposes a thermodynamic formalism for stationary nonlinear resistive networks, centered on generalized transport potentials (GTPs) Phi(J)=int X(J')dJ' and Psi(X)=int J(X')dX'. It claims that extremizing the Lagrangians L_J=XJ-Phi(J), L_X=XJ-Psi(X), and the Gyarmati combination unifies known extremum principles for dissipation and entropy production, that stability changes among multiple steady states can be interpreted as nonequilibrium phase transitions, and that optimizing branch resistances with maintenance costs (Murray-type optimization) produces apparent transport laws whose convexity determines loop versus tree network structures, thereby connecting to branched optimal transport. The paper also applies the stability criterion to a self-organized plasma boundary-layer model.

Significance. If the claims were all correct, the paper would provide a useful unifying language: the power-law case would recover maximum/minimum entropy production and minimum dissipation as special cases, the plasma example would show how GTP-based criteria outperform dissipation-based criteria, and the Murray-cost optimization would be placed in a common framework with branched optimal transport. The manuscript is clearly written, candid about limitations, and careful to separate generalized dissipation from entropy production. The plasma-model stability analysis is concrete and internally consistent. However, the generality of the Section 7 results is undermined by an algebraic error in the apparent-resistance exponent and an internal contradiction about which regime yields loops and which yields trees. In addition, the foundational variational principle is partly definitional, and the paper should state more explicitly which conclusions depend on an assumed relaxation dynamics. These issues are localized and correctable, so the manuscript merits major revision rather than rejection.

major comments (4)
  1. [Section 7, Eq. (49)] The apparent resistance exponent is algebraically wrong for general power laws. Substituting the optimized resistance from Eq. (48), Re_i=(k beta)^{1/(beta+1)} J_i^{-(1+alpha)/(1+beta)}, into the force-current law X_i=Re_i J_i^alpha gives X_i proportional to J_i^{alpha-(1+alpha)/(1+beta)} = J_i^{(alpha beta-1)/(1+beta)}. Equation (49) instead states chec_alpha=(beta-alpha)/(1+beta). The two expressions agree only when alpha=1. Since the loop/tree classification in the following paragraphs depends on the value and sign of the apparent exponent, the classification for power-law resistances with alpha not equal to 1 is not established as stated.
  2. [Section 7, Eq. (50)] The optimized-cost coefficient in Eq. (50) is also incorrect. Substituting Eq. (48) into Eq. (46) gives M_i^*=(beta k)^{1/(beta+1)} (1+1/beta) J_i^{beta(alpha+1)/(beta+1)}, not (beta k)^{1/(beta+1)} (1+(beta k)^beta) J_i^{beta(alpha+1)/(beta+1)}. The printed expression is dimensionally inconsistent unless some normalization makes (beta k)^beta dimensionless. This indicates that the reduction to the effective cost functional was not carried out consistently and should be redone.
  3. [Section 7, final two paragraphs after Eq. (51)] The manuscript contradicts itself about which regime gives loops and which gives trees. The paragraph ending with "resulting in tree networks (corresponding to the unstable transport law with chec_alpha > 1)" states that loops appear for chec_alpha<1 and trees for chec_alpha>1. The very next paragraph says the opposite: "For chec_alpha>1, the functional is convex with a single minimum corresponding to the most balanced network with loops, while for chec_alpha<1 it develops multiple singular minima in correspondence of tree networks." The connection to branched optimal transport hinges on this dichotomy, so the authors should identify the correct condition, correct the text, and make sure Figure 7 is consistent with it.
  4. [Sections 2 and 6, Eqs. (7)-(8) and Eq. (33)] The variational principle is, in an important sense, definitional: because Phi(J) is defined as the integral of the constitutive law X(J), the stationarity condition of L_J with respect to J is exactly X=R(J)J. The paper acknowledges this by allowing virtual variations that do not satisfy the transport law, but the stability analysis in Sec. 6 goes beyond this formal identity. In particular, Eq. (33) postulates a gradient relaxation dJ/dt=partial L_J/partial J, and the stability conclusions of Secs. 6.1 and 6.2 follow from that dynamical assumption, not from the variational construction alone. The authors should explicitly separate the formal Legendre-transform identity from the physically assumed relaxation dynamics and state which conclusions require the latter.
minor comments (3)
  1. [Section 2, paragraph after Fig. 1] The text writes "the GPTs are no longer proportional to dissipation"; this should be "GTPs" for consistency with the rest of the paper.
  2. [Section 6, Eqs. (39) and (44)] The relaxation equations mix variables: Eq. (39) has tau_X dX/dt on the left but a derivative with respect to J on the right, and similarly Eq. (44) should be checked for consistency. Please clarify the intended dynamical variables in these equations.
  3. [Section 7, sentence before Eq. (50)] The phrase "Further including either the condition (48)" is unclear; it should probably read "Further including the condition (48)" or "Eliminating the resistances using (48)".

Circularity Check

2 steps flagged · score 7.0 of 10

The central variational principle is self-definitional: the Lagrangian's stationarity is the constitutive law by construction, and the stability criterion inherits the same construction; Section 7 separately contains an algebraic slip.

  1. self definitional [Sec. 2, Eqs. (7)-(8) and following text]
    "the optimization principle is based on the extremization of the Lagrangian [14] LJ (J) = XJ − Φ(J), (7) where the GTP (or content) [12] is Φ(J) = Z J 0 X(J ′)dJ ′. (8) ... The system’s stable operating point (2) is obtained for the values of the current that maximizes LJ ."

    By the fundamental theorem of calculus, dΦ/dJ = X(J). Hence the stationarity condition dLJ/dJ = X − Φ′(J) = X − X(J) = 0 is exactly the constitutive law (2). The 'operating point extremum' is therefore a restatement of the X(J) used to define Φ; it is not an independent variational derivation. The paper's own setup allows perturbations 'which do not necessarily satisfy the basic transport law (2)', but the extremum condition restores that law by construction, so the optimality principle carries no information beyond the input constitutive law.

  2. self definitional [Sec. 6, Eq. (33) and stability discussion]
    "For a flux-driven 1D system, the unsteady evolution around the operating point may be described formally by an evolution equation of the type dJ dt = ∂LJ (J) ∂J = X − ∂Φ(J) ∂J = X − R(J)J, (33) ... the stability and selection of the NESS should be based on the Lagrangians (7) and (9) and their second variations."

    Since Φ(J) is defined as the integral of X(J), the second derivative of LJ is −dX/dJ. Labeling maxima 'stable' and minima 'unstable' is then equivalent to assuming gradient descent on a potential whose derivative is the constitutive law; the stability classification restates the slope of X(J) under an assumed dynamics rather than predicting it. The phrase 'may be described formally' concedes that the evolution equation is assumed, so the phase-transition interpretation is built into the definition of LJ.

full rationale

The central variational statement of Sec. 2 is a definitional identity: L_J is built from Φ(J)=∫ X(J')dJ', so the Euler-Lagrange condition is the constitutive law itself. This is the paper's own account and is not a prediction from independent first principles. The same construction drives the Sec. 6 stability criterion, because the assumed gradient flow's Hessian is just the derivative of the input law. These are genuine instances of self-definitional circularity in the core formalism. However, the paper is not merely a self-citation chain: the power-law network reduction in Sec. 5 (dissipation minimization from KCL), the multi-branch extension, and the Murray-cost optimization in Sec. 7 are independent algebraic calculations, and the convexity/loop-tree classification is attributed to non-author references [76] and [4]. Self-citations such as [23], [44], [74], and [75] are contextual or supporting rather than load-bearing. Separately from circularity, Eq. (49) appears algebraically inconsistent for general α: substituting Eq. (48) into X=ℜJ^α gives exponent (αβ−1)/(1+β), not (β−α)/(1+β); the two coincide only for α=1. This is a correctness defect, not a circularity, but it weakens the claimed general connection to branched optimal transport. Overall, because the paper's foundational 'extremization selects the operating point' claim reduces to the definition of the potential, the circularity score is 7, with the caveat that substantial sections retain independent content.

Assumptions & free parameters 3 free parameters · 6 assumptions · 0 invented entities

The paper is a theoretical synthesis with no data fitting. The main inputs are the constitutive family and the cost family; the parameters alpha, beta, and k are model inputs rather than fitted values. The variational principle is definitional: the potential is the integral of the force law. Stability and network conclusions then follow from convexity/concavity arguments on these potentials.

free parameters (3)
  • alpha (power-law resistance exponent) = not fitted; model parameter
    Defines the resistance family R = ℜ J^(alpha-1) in Eq. (3). Many results, such as proportionality of GTP to dissipation and the entropy-production equivalence at alpha=1, depend on alpha, but alpha is treated as an input, not tuned to data.
  • beta (maintenance cost exponent) = not fitted; model parameter
    Defines the cost k / ℜ^beta in Eq. (46). The sign of the apparent resistance exponent and the loop/tree transition depend on beta, but beta is not fitted to data.
  • k (cost coefficient) = not fitted; model parameter
    Appears in the cost Eq. (46). It sets the optimal resistance scale but cancels in the exponent; not tuned to data.
assumptions (6)
  • domain assumption Steady-state, single-energy-mode network with no cross-coupling between fluxes and forces and no storage terms in branches.
    Introduced in the first bullet of Sec. 1; excludes Onsager coupling and time-dependent storage, so all later extremization results are confined to this class.
  • domain assumption The constitutive law depends only on the local force/current, not on the absolute value of the driving potential.
    Stated in Sec. 1 bullet; needed for the potentials Phi(J) and Psi(X) to be well-defined functions of local variables.
  • ad hoc to paper Virtual variations around the operating point are allowed even when they violate the constitutive law, and the extremum of the resulting Lagrangian selects the physical operating point.
    Introduced in Sec. 2 before Eq. (7); this is the load-bearing interpretational premise that turns the identity dPhi/dJ = X(J) into an optimality principle.
  • domain assumption Power-law resistance family R = ℜ J^(alpha-1) is used for the entropy-production and dissipation equivalence results.
    Eq. (3) and Sec. 5 use this family to make Phi proportional to dissipation; for general nonlinear laws the entropy-production equivalences fail.
  • domain assumption Isothermal conditions with uniform T in each branch are required for entropy production to be proportional to dissipation.
    Eq. (6) and Secs. 1 and 5; the paper stresses entropy-production extremization is misleading outside isothermal power-law cases.
  • domain assumption Cost function for optimizing transport properties has the power-law form M = ℜ J^(alpha+1) + k / ℜ^beta.
    Eq. (46) in Sec. 7; the loop/tree conclusions and the Murray exponent follow from this specific cost family.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Thermodynamics and Legendre Duality in Optimal Networks." pith.science (2026). https://pith.science/paper/BBGPN4SX

@misc{pith2026250615727,
  author       = {Pith},
  title        = {Pith review of: Thermodynamics and Legendre Duality in Optimal Networks},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BBGPN4SX}},
  note         = {Machine review of arXiv:2506.15727}
}
read the original abstract

Optimality principles in nonequilibrium transport networks are linked to a thermodynamic formalism based on generalized transport potentials endowed with Legendre duality and related contact structure. This allows quantifying the distance from non-equilibrium operating points, analogously to thermodynamic availability as well as to shed light on optimality principles in relation to different imposed constraints. Extremizations of generalized dissipation and entropy production appear as special cases that require power-law resistances and -- for entropy production -- also isothermal conditions. Changes in stability of multiple operating points are interpreted as phase transitions based on non-equilibrium equations of state, while cost-based optimization of transport properties reveals connections to the generalized dissipation in the case of power law costs and linear resistance law, but now with typically unstable operating points which give rise to branched optimal transport.

Figures

Figures reproduced from arXiv: 2506.15727 by the authors.

Figure 1
Figure 1. GTPs (content and co-content) for nonlinear transport law, [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗
Figure 2
Figure 2. Contour plot of the Lagrangian LJ (7) for power law resistance (3) as a function of current J and power-law exponent α. Lighter colors mean higher values of the Lagrangian. We have already mentioned that the principle obtained by maximizing the Lagrangian LJ , Eq. (7), can be seen as a generalization of the principles of Rayleigh and Onsager to nonlinear laws, although for generalized dissipation rates, rather than … view at source ↗
Figure 3
Figure 3. Under exponential resistance law, R(J) = e J − 1; despite the difference between LJ (blue curve) and LJλ (orange curve), their maximum is at the same value of the flux. 8 [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Parallel pipes. Contour plots for the complete Lagrangians [PITH_FULL_IMAGE:figures/full_fig_p012_4.png]
Figure 5
Figure 5. Figure 5: Self-organized boundary layer in a force-driven plasma flow. (a) Dissipation for the linear (blue) [PITH_FULL_IMAGE:figures/full_fig_p016_5.png]
Figure 6
Figure 6. Figure 6: Self-organized boundary layer in a flux-driven plasma flow. (a) Dissipation for the linear (blue) [PITH_FULL_IMAGE:figures/full_fig_p017_6.png]
Figure 7
Figure 7. Figure 7: Selection of optimal network configurations for the triple loop case, with unit input in the top left [PITH_FULL_IMAGE:figures/full_fig_p019_7.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

80 extracted references · 80 canonical work pages

  1. [1]

    R. B. Braithwaite, et al., Scientific Explanation a Study of the Function of Theory, Probability and Law in Science, CUP Archive, 1953

    work page 1953

  2. [2]

    B. J. Chung, B. De Bari, J. Dixon, D. Kondepudi, J. Pateras, A. Vaidya, On the thermodynamics of self-organization in dissipative systems: Reflections on the unification of physics and biology, Fluids 7 (4) (2022) 141

    work page 2022

  3. [3]

    M. J. Sewell, Maximum and minimum principles: a unified approach with applications, Vol. 1, CUP Archive, 1987

    work page 1987

  4. [4]

    Rodriguez-Iturbe, A

    I. Rodriguez-Iturbe, A. Rinaldo, Fractal river basins: chance and self-organization, Cambridge Univer- sity Press, 1997

    work page 1997

  5. [5]

    Santambrogio, Optimal transport for applied mathematicians, Birk¨ auser, NY 55 (58-63) (2015) 94

    F. Santambrogio, Optimal transport for applied mathematicians, Birk¨ auser, NY 55 (58-63) (2015) 94

    work page 2015

  6. [6]

    Van Baak, Variational alternatives to kirchhoff’s loop theorem in dc circuits, American Journal of Physics 67 (1) (1999) 36–44

    D. Van Baak, Variational alternatives to kirchhoff’s loop theorem in dc circuits, American Journal of Physics 67 (1) (1999) 36–44

    work page 1999

  7. [7]

    R. K. Niven, Steady state of a dissipative flow-controlled system and the maximum entropy production principle, Physical Review E 80 (2) (2009) 021113

    work page 2009

  8. [8]

    H. B. Callen, Thermodynamics and an Introduction to Thermostatistics, John Wiley & Sons, 1998

    work page 1998

Show all 80 references
  1. [9]

    Barbera, On the principle of minimal entropy production for navier-stokes-fourier fluids, Continuum Mechanics and Thermodynamics 11 (5) (1999) 327–330

    E. Barbera, On the principle of minimal entropy production for navier-stokes-fourier fluids, Continuum Mechanics and Thermodynamics 11 (5) (1999) 327–330

    1999

  2. [10]

    Kondepudi, I

    D. Kondepudi, I. Prigogine, Modern thermodynamics: from heat engines to dissipative structures, John Wiley & Sons, 2014

    2014

  3. [11]

    Martyushev, A

    L. Martyushev, A. Nazarova, V. Seleznev, On the problem of the minimum entropy production in the nonequilibrium stationary state, Journal of Physics A: Mathematical and Theoretical 40 (3) (2006) 371

    2006

  4. [12]

    Millar, Cxvi

    W. Millar, Cxvi. some general theorems for non-linear systems possessing resistance, The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science 42 (333) (1951) 1150–1160

    1951

  5. [13]

    Cherry, Non-linear circuit theory by the methods of classical dynamics, in: Recent Developments in Network Theory, Elsevier, 1963, pp

    C. Cherry, Non-linear circuit theory by the methods of classical dynamics, in: Recent Developments in Network Theory, Elsevier, 1963, pp. 211–219

    1963

  6. [14]

    Verh´ as, Gyarmati’s variational principle of dissipative processes, Entropy 16 (4) (2014) 2362–2383

    J. Verh´ as, Gyarmati’s variational principle of dissipative processes, Entropy 16 (4) (2014) 2362–2383

    2014

  7. [15]

    C. D. Murray, The physiological principle of minimum work: I. the vascular system and the cost of blood volume, Proceedings of the National Academy of Sciences 12 (3) (1926) 207–214

    1926

  8. [16]

    J. R. Banavar, A. Maritan, A. Rinaldo, Size and form in efficient transportation networks, Nature 399 (6732) (1999) 130–132

    1999

  9. [17]

    Bejan, S

    A. Bejan, S. Lorente, The constructal law and the thermodynamics of flow systems with configuration, International journal of heat and mass transfer 47 (14-16) (2004) 3203–3214

    2004

  10. [18]

    A. Tero, R. Kobayashi, T. Nakagaki, A mathematical model for adaptive transport network in path finding by true slime mold, Journal of theoretical biology 244 (4) (2007) 553–564

    2007

  11. [19]

    S. Bohn, M. O. Magnasco, Structure, scaling, and phase transition in the optimal transport network, Physical review letters 98 (8) (2007) 088702

    2007

  12. [20]

    Rinaldo, R

    A. Rinaldo, R. Rigon, J. R. Banavar, A. Maritan, I. Rodriguez-Iturbe, Evolution and selection of river networks: Statics, dynamics, and complexity, Proceedings of the National Academy of Sciences 111 (7) (2014) 2417–2424. 21

    2014

  13. [21]

    Bandle, Dido’s problem and its impact on modern mathematics, Notices of the AMS 64 (9)

    C. Bandle, Dido’s problem and its impact on modern mathematics, Notices of the AMS 64 (9)

  14. [22]

    Hooshyar, S

    M. Hooshyar, S. Anand, A. Porporato, Variational analysis of landscape elevation and drainage net- works, Proceedings of the Royal Society A 476 (2239) (2020) 20190775

    2020

  15. [23]

    S. K. Anand, M. B. Bertagni, T. D. Drivas, A. Porporato, Self-similarity and vanishing diffusion in fluvial landscapes, Proceedings of the National Academy of Sciences 120 (51) (2023) e2302401120

    2023

  16. [24]

    Facca, F

    E. Facca, F. Cardin, M. Putti, Towards a stationary monge–kantorovich dynamics: The physarum polycephalum experience, SIAM Journal on Applied Mathematics 78 (2) (2018) 651–676

    2018

  17. [25]

    Mrugala, J

    R. Mrugala, J. D. Nulton, J. C. Sch¨ on, P. Salamon, Contact structure in thermodynamic theory, Reports on mathematical physics 29 (1) (1991) 109–121

    1991

  18. [26]

    M. A. Nowak, Evolutionary dynamics: exploring the equations of life, Harvard university press, 2006

    2006

  19. [27]

    Schnakenberg, Network theory of microscopic and macroscopic behavior of master equation systems, Reviews of Modern physics 48 (4) (1976) 571

    J. Schnakenberg, Network theory of microscopic and macroscopic behavior of master equation systems, Reviews of Modern physics 48 (4) (1976) 571

    1976

  20. [28]

    Oster, C

    G. Oster, C. Desoer, Tellegen’s theorem and thermodynamic inequalities, Journal of theoretical Biology 32 (2) (1971) 219–241

    1971

  21. [29]

    L. M. Martyushev, V. D. Seleznev, Maximum entropy production principle in physics, chemistry and biology, Physics Reports 426 (1) (2006) 1–45

    2006

  22. [30]

    E. T. Jaynes, The minimum entropy production principle, Annual Review of Physical Chemistry 31 (1) (1980) 579–601

    1980

  23. [31]

    Onsager, S

    L. Onsager, S. Machlup, Fluctuations and irreversible processes, Physical Review 91 (6) (1953) 1505

    1953

  24. [32]

    J. W. S. B. Rayleigh, Some general theorems relating to vibrations, Vol. 4, London Mathematical Society, 1873

  25. [33]

    Doi, Onsager’s variational principle in soft matter, Journal of Physics: Condensed Matter 23 (28) (2011) 284118

    M. Doi, Onsager’s variational principle in soft matter, Journal of Physics: Condensed Matter 23 (28) (2011) 284118

    2011

  26. [34]

    Bataille, D

    J. Bataille, D. Edelen, J. Kestin, On the structuring of thermodynamic fluxes: a direct implementation of the dissipation inequality, International Journal of Engineering Science 17 (5) (1979) 563–572

    1979

  27. [35]

    H. L. Langhaar, The principle of complementary energy in nonlinear elasticity theory, Journal of the Franklin Institute 256 (3) (1953) 255–264

    1953

  28. [36]

    Sewell, Complementary energy and catastrophes, in: Variational Methods in the Mechanics of Solids, Elsevier, 1980, pp

    M. Sewell, Complementary energy and catastrophes, in: Variational Methods in the Mechanics of Solids, Elsevier, 1980, pp. 163–168

    1980

  29. [37]

    Gyarmati, et al., Non-equilibrium thermodynamics, Vol

    I. Gyarmati, et al., Non-equilibrium thermodynamics, Vol. 184, Springer, 1970

    1970

  30. [38]

    Biolek, D

    Z. Biolek, D. Biolek, V. Biolkov´ a, (co) content in circuits with memristive elements, IEEE Transactions on Circuits and Systems I: Regular Papers 62 (2) (2014) 488–496

    2014

  31. [39]

    Ziegler, An introduction to thermomechanics, Elsevier, 2012

    H. Ziegler, An introduction to thermomechanics, Elsevier, 2012

    2012

  32. [40]

    Gallavotti, Statistical mechanics: A short treatise, Springer Science & Business Media, 2013

    G. Gallavotti, Statistical mechanics: A short treatise, Springer Science & Business Media, 2013

    2013

  33. [41]

    ˇZupanovi´ c, D

    P. ˇZupanovi´ c, D. Jureti´ c, S. Botri´ c, Kirchhoff’s loop law and the maximum entropy production principle, Physical Review E 70 (5) (2004) 056108

    2004

  34. [42]

    D. G. Edelen, A nonlinear onsager theory of irreversibility, International Journal of Engineering Science 10 (6) (1972) 481–490

    1972

  35. [43]

    L. D. Landau, E. M. Lifshitz, Statistical Physics: Volume 5, Vol. 5, Elsevier, 2013

    2013

  36. [44]

    Porporato, S

    A. Porporato, S. Calabrese, L. Rondoni, Thermodynamic langevin equations, Physical Review E 110 (5) (2024) 054136

    2024

  37. [45]

    Schl¨ ogl, Thermodynamic metric and stochastic measures, Zeitschrift f¨ ur Physik B Condensed Matter 59 (4) (1985) 449–454

    F. Schl¨ ogl, Thermodynamic metric and stochastic measures, Zeitschrift f¨ ur Physik B Condensed Matter 59 (4) (1985) 449–454

    1985

  38. [46]

    Mrugala, J

    R. Mrugala, J. D. Nulton, J. C. Sch¨ on, P. Salamon, Statistical approach to the geometric structure of thermodynamics, Physical Review A 41 (6) (1990) 3156

    1990

  39. [47]

    Brody, N

    D. Brody, N. Rivier, Geometrical aspects of statistical mechanics, Physical Review E 51 (2) (1995) 1006

    1995

  40. [48]

    G. E. Crooks, Measuring thermodynamic length, Physical Review Letters 99 (10) (2007) 100602

    2007

  41. [49]

    R. K. Niven, B. Andresen, Jaynes’ maximum entropy principle, riemannian metrics and generalised least action bound, in: Complex Physical, Biophysical and Econophysical Systems, World Scientific, 2010, pp. 283–317. 22

    2010

  42. [50]

    Ercan, On kirchhoff’s equations and variational approaches to electrical network analysis, American Journal of Physics 84 (3) (2016) 231–233

    A. Ercan, On kirchhoff’s equations and variational approaches to electrical network analysis, American Journal of Physics 84 (3) (2016) 231–233

    2016

  43. [51]

    R. F. Greene, H. B. Callen, On the formalism of thermodynamic fluctuation theory, Physical Review 83 (6) (1951) 1231

    1951

  44. [52]

    R. K. Niven, Simultaneous extrema in the entropy production for steady-state fluid flow in parallel pipes, J. Non-Equilib. Thermodyn. 35 (2010) 347–378

    2010

  45. [53]

    N. S. Harsha, A review of the variational methods for solving dc circuits, European Journal of Physics 40 (3) (2019) 033001

    2019

  46. [54]

    Kawazura, Z

    Y. Kawazura, Z. Yoshida, Comparison of entropy production rates in two different types of self- organized flows: B´ enard convection and zonal flow, Physics of Plasmas 19 (1) (2012) 012305

    2012

  47. [55]

    Martyushev, Some interesting consequences of the maximum entropy production principle, Journal of Experimental and Theoretical Physics 104 (2007) 651–654

    L. Martyushev, Some interesting consequences of the maximum entropy production principle, Journal of Experimental and Theoretical Physics 104 (2007) 651–654

    2007

  48. [56]

    Hill, Entropy production as the selection rule between different growth morphologies, Nature 348 (1990) 426–428

    A. Hill, Entropy production as the selection rule between different growth morphologies, Nature 348 (1990) 426–428

    1990

  49. [57]

    Ozawa, S

    H. Ozawa, S. Shimokawa, H. Sakuma, Thermodynamics of fluid turbulence: A unified approach to the maximum transport properties, Physical Review E 64 (2) (2001) 026303

    2001

  50. [58]

    J. J. Vallino, Ecosystem biogeochemistry considered as a distributed metabolic network ordered by maximum entropy production, Philosophical Transactions of the Royal Society B: Biological Sciences 365 (1545) (2010) 1417–1427

    2010

  51. [59]

    J. J. Vallino, C. K. Algar, The thermodynamics of marine biogeochemical cycles: Lotka revisited, Annual Review of Marine Science 8 (2016) 333–356

    2016

  52. [60]

    H. Ge, H. Qian, Thermodynamic limit of a nonequilibrium steady state: Maxwell-type construction for a bistable biochemical system, Physical review letters 103 (14) (2009) 148103

    2009

  53. [61]

    Yoshida, S

    Z. Yoshida, S. Mahajan, “maximum” entropy production in self-organized plasma boundary layer: A thermodynamic discussion about turbulent heat transport, Physics of Plasmas 15 (3) (2008) 032307

    2008

  54. [62]

    Yoshida, Y

    Z. Yoshida, Y. Kawazura, Bifurcation, stability, and entropy production in a self-organizing fluid/plasma system, Beyond the Second Law: Entropy Production and Non-equilibrium Systems (2014) 291–307

    2014

  55. [63]

    Kawazura, Z

    Y. Kawazura, Z. Yoshida, Entropy production rate in a flux-driven self-organizing system, Physical Review E 82 (6) (2010) 066403

    2010

  56. [64]

    Glansdorff, I

    P. Glansdorff, I. Prigogine, On a general evolution criterion in macroscopic physics, Physica 30 (2) (1964) 351–374

    1964

  57. [65]

    Landauer, Stability and entropy production in electrical circuits, Journal of Statistical Physics 13 (1975) 1–16

    R. Landauer, Stability and entropy production in electrical circuits, Journal of Statistical Physics 13 (1975) 1–16

    1975

  58. [66]

    Bodineau, B

    T. Bodineau, B. Derrida, A perturbative approach to the macroscopic fluctuation theory, Journal of Statistical Physics 192 (4) (2025) 1–20

    2025

  59. [67]

    Derrida, J

    B. Derrida, J. Lebowitz, E. Speer, Free energy functional for nonequilibrium systems: an exactly solvable case, Physical Review Letters 87 (15) (2001) 150601

    2001

  60. [68]

    Bertini, A

    L. Bertini, A. De Sole, D. Gabrielli, G. Jona-Lasinio, C. Landim, Macroscopic fluctuation theory for stationary non-equilibrium states, Journal of Statistical Physics 107 (2002) 635–675

    2002

  61. [69]

    Torquato, Optimal design of heterogeneous materials, Annual review of materials research 40 (1) (2010) 101–129

    S. Torquato, Optimal design of heterogeneous materials, Annual review of materials research 40 (1) (2010) 101–129

    2010

  62. [70]

    Waldherr, D

    S. Waldherr, D. A. Oyarz´ un, A. Bockmayr, Dynamic optimization of metabolic networks coupled with gene expression, Journal of theoretical biology 365 (2015) 469–485

    2015

  63. [71]

    Franklin, S

    O. Franklin, S. P. Harrison, R. Dewar, C. E. Farrior, ˚A. Br¨ annstr¨ om, U. Dieckmann, S. Pietsch, D. Falster, W. Cramer, M. Loreau, et al., Organizing principles for vegetation dynamics, Nature plants 6 (5) (2020) 444–453

    2020

  64. [72]

    Calabrese, B

    S. Calabrese, B. P. Mohanty, A. A. Malik, Soil microorganisms regulate extracellular enzyme production to maximize their growth rate, Biogeochemistry 158 (3) (2022) 303–312

    2022

  65. [73]

    Witelski, M

    T. Witelski, M. Bowen, Methods of mathematical modelling, Springer, 2015

    2015

  66. [74]

    S. K. Anand, M. Hooshyar, J. Martin Nordbotten, A. Porporato, A minimalist model for coevolving 23 supply and drainage networks, Royal Society Open Science 8 (2) (2021) 201407

    2021

  67. [75]

    Giaccone, S

    L. Giaccone, S. K. Anand, A. Porporato, L. Ridolfi, Coexistence of loop and tree structures in transport networks, Physical Review E 111 (1) (2025) 014319

    2025

  68. [76]

    J. R. Banavar, F. Colaiori, A. Flammini, A. Maritan, A. Rinaldo, Scaling, optimality, and landscape evolution, Journal of Statistical Physics 104 (2001) 1–48

    2001

  69. [77]

    Burger, J

    M. Burger, J. Haskovec, P. A. Markowich, A mesoscopic model of biological transportation networks, Communications in Mathematical Sciences 17 (5) 1213–1234

  70. [78]

    D. Hu, D. Cai, Adaptation and optimization of biological transport networks, Physical review letters 111 (13) (2013) 138701

    2013

  71. [79]

    Bonetti, M

    S. Bonetti, M. Hooshyar, C. Camporeale, A. Porporato, Channelization cascade in landscape evolution, Proceedings of the National Academy of Sciences 117 (3) (2020) 1375–1382

    2020

  72. [80]

    Maritan, F

    A. Maritan, F. Colaiori, A. Flammini, M. Cieplak, J. R. Banavar, Universality classes of optimal channel networks, Science 272 (5264) (1996) 984–986. 24

    1996

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.