pith. sign in

arxiv: 2607.00584 · v1 · pith:LMCX3MMInew · submitted 2026-07-01 · 🧬 q-bio.PE

Optimal control on a heterogeneous SI epidemic model

Elisa Paparelli (SU) This is my paper

Pith reviewed 2026-07-02 02:00 UTC · model grok-4.3

classification 🧬 q-bio.PE
keywords optimal controlPontryagin Minimum PrincipleSI epidemic modelheterogeneityfinal sizesupply constraintmacroscopic reduction
0
0 comments X

The pith

The Pontryagin Minimum Principle characterizes an optimal pharmaceutical control that minimizes final infection size in a heterogeneous SI model under an integral supply constraint.

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

This paper sets up an optimal control problem on the macroscopic equations of an SI epidemic model that already incorporates population-level heterogeneity in resistance and viral load. The goal is to choose a control function that reduces the total number of infected individuals at a fixed final time while obeying a hard limit on total supply expressed as an integral equality constraint. Application of the Pontryagin Minimum Principle supplies the necessary conditions that any such minimizing control must satisfy. A sympathetic reader would care because the same heterogeneity that makes real populations hard to model also makes uniform control strategies inefficient, so an optimality condition derived on the reduced model offers a concrete way to allocate limited interventions.

Core claim

For the macroscopic SI dynamics obtained from the heterogeneous framework, the Pontryagin Minimum Principle yields a characterization of the optimal control that minimizes the final size of the infection at a prescribed terminal time subject to the integral equality constraint on the control.

What carries the argument

Pontryagin Minimum Principle applied to the reduced macroscopic SI system, which supplies the adjoint equations and the pointwise minimization condition that determine the candidate optimal control.

If this is right

  • Any optimal control must satisfy the pointwise minimization condition given by the Hamiltonian evaluated along the adjoint trajectory.
  • The final infection size achieved by the optimal control is the smallest possible value consistent with the total supply limit and the macroscopic dynamics.
  • The adjoint system couples the state and costate variables so that the control depends on both current prevalence and the shadow price of future infections.
  • The integral constraint is handled by introducing a constant Lagrange multiplier that appears in the Hamiltonian.

Where Pith is reading between the lines

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

  • The same reduction-to-macroscopic-dynamics step could be repeated for other compartmental models once their heterogeneous counterparts are available.
  • If the macroscopic model is solved numerically with the derived optimality conditions, the resulting control trajectory supplies a testable prediction for how limited pharmaceutical resources should be timed.
  • Heterogeneity parameters that enter the macroscopic equations become explicit levers that could be varied to study how more or less variable populations change the shape of the optimal control.

Load-bearing premise

The macroscopic equations derived from the heterogeneous individual-level model are accurate enough to serve as the state dynamics for the optimal control problem.

What would settle it

A numerical simulation of the full heterogeneous agent-based model in which the control obtained from the macroscopic optimality conditions produces a strictly larger final infection size than a feasible alternative strategy would falsify the claim that the macroscopic reduction preserves the optimum.

Figures

Figures reproduced from arXiv: 2607.00584 by Elisa Paparelli (SU).

Figure 1
Figure 1. Figure 1: Optimal solution under assumptions of Theorems (1) and (2). Optimal trajectories of NS, NI , MS (the mean of resistance level in susceptible compartment, i.e MS = PS/NS and MI (the mean viral load in infected population, i.e. MI = PI/NI ) obtained by solving numerically the optimal control problem (15) (continuous line) and the individual-based dynamics (1)-(6) (circular markers) inserting the optimal cont… view at source ↗
Figure 2
Figure 2. Figure 2: Optimal solution extending Theorem 1 and the uncontrolled dynamics. On the right the trajectories of the functions NS, NI , MS, and MI without any control, obtained by prescribing q defined via (84). On the left, the optimal trajectories obtained by implementing the corresponding optimal control problem, extending the time horizon. 19 [PITH_FULL_IMAGE:figures/full_fig_p019_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Optimal solution extending Theorem 2 and the uncontrolled dynamics. On the right the trajectories of the functions NS, NI , MS, and MI without any control, obtained by prescribing q defined via (85). On the left, the optimal trajectories obtained by implementing the corresponding optimal control problem, extending the time horizon. References [1] T. Lorenzi, E. Paparelli, A. Tosin, Modelling coevolutionary… view at source ↗
read the original abstract

This work addresses an optimal control problem for a SI epidemic model incorporating heterogeneities in resistance and viral load at the population level. Building upon the heterogeneous SI framework developed in [1], a minimization problem constrained to the macroscopic counterpart of the SI dynamics derived therein is proposed. Unlike traditional optimal control problems in homogeneous epidemic models, the present approach focuses on an optimal control problem that accounts for population heterogeneity, offering insights from a microscale perspective. The contribution aims to minimize the final size of the infection within a finite time horizon by developing a pharmaceutical strategy, under a supply constraint that translates into an integral equality constraint in the control function. By applying the Pontryagin Minimum Principle, a characterization of an optimal control is provided.

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 formulates an optimal control problem on the macroscopic SI dynamics taken from the heterogeneous framework of reference [1]. The goal is to minimize final infection size over a finite horizon subject to an integral supply constraint on a pharmaceutical control; the Pontryagin Minimum Principle is invoked to obtain a characterization of the optimal control.

Significance. If the macroscopic reduction remains faithful once the control is introduced, the work supplies a concrete PMP-based characterization that incorporates heterogeneity in resistance and viral load. The approach is a direct extension of existing optimal-control techniques to a reduced heterogeneous model, but its practical value hinges on the unverified compatibility of the control with the original microscale reduction.

major comments (2)
  1. [Model reduction and control formulation] The derivation of the macroscopic equations and the manner in which the control enters the microscale model are not shown to be compatible with the reduction performed in [1]. Because the optimal-control characterization is obtained exclusively on the reduced system, this compatibility is load-bearing for the claim that the resulting strategy minimizes final size in the heterogeneous population the problem is intended to address.
  2. [Numerical verification or results] No numerical experiment is reported that re-embeds the macro-optimal control into the original heterogeneous agent-based or network model and verifies that the final-size reduction is preserved. Without this check the central claim that the PMP characterization solves the intended heterogeneous problem cannot be assessed.
minor comments (2)
  1. [Abstract and §3] The abstract states that PMP yields a characterization but supplies neither the explicit Hamiltonian nor the boundary conditions; these should be stated in the main text with equation numbers.
  2. [Notation section] Notation for the macroscopic state variables and the control constraint should be introduced once and used consistently; several symbols appear to be carried over from [1] without re-definition.

Simulated Author's Rebuttal

2 responses · 0 unresolved

Thank you for the opportunity to respond to the referee's comments on our manuscript. We address each major comment below with clarifications on the scope of our work and indicate revisions where appropriate.

read point-by-point responses

  1. Referee: [Model reduction and control formulation] The derivation of the macroscopic equations and the manner in which the control enters the microscale model are not shown to be compatible with the reduction performed in [1]. Because the optimal-control characterization is obtained exclusively on the reduced system, this compatibility is load-bearing for the claim that the resulting strategy minimizes final size in the heterogeneous population the problem is intended to address.

    Authors: The macroscopic equations are taken directly from the reduction derived in [1] for the uncontrolled heterogeneous SI model. The pharmaceutical control is incorporated at the macroscopic level as a time-dependent adjustment to the effective transmission rate, which aligns with the population-level effect of such an intervention. We agree that an explicit argument showing how the control would enter the underlying microscale dynamics while preserving the reduction is needed to fully justify the claim. In the revised version we will add a subsection that outlines this compatibility assumption based on the structure of the reduction in [1] and discusses the conditions under which it remains valid. revision_made = 'yes' revision: yes

  2. Referee: [Numerical verification or results] No numerical experiment is reported that re-embeds the macro-optimal control into the original heterogeneous agent-based or network model and verifies that the final-size reduction is preserved. Without this check the central claim that the PMP characterization solves the intended heterogeneous problem cannot be assessed.

    Authors: Our contribution is the analytic characterization of the optimal control via the PMP on the macroscopic reduced system, which is presented as a faithful proxy for the heterogeneous population. We acknowledge that direct numerical re-embedding into the original agent-based or network model would provide additional empirical support for the final-size reduction. Such verification, however, requires substantial new computational work outside the theoretical scope of the present manuscript. In revision we will add an explicit statement clarifying that the results apply to the macroscopic model and note the desirability of future microscale validation. revision_made = 'partial' revision: partial

Circularity Check

1 steps flagged

Optimal control result depends on macroscopic reduction imported from self-cited prior work [1]

specific steps
  1. self citation load bearing [Abstract]
    "Building upon the heterogeneous SI framework developed in [1], a minimization problem constrained to the macroscopic counterpart of the SI dynamics derived therein is proposed."

    The minimization problem and subsequent PMP characterization are defined directly on the macroscopic dynamics taken from [1]; the well-posedness of the optimal control problem therefore rests on the accuracy of that prior reduction, which is not re-derived or externally validated here.

full rationale

The paper's core contribution is the application of the Pontryagin Minimum Principle to characterize an optimal control on the given dynamics. This PMP step is independent and adds new content. However, the load-bearing macroscopic SI model is taken directly from reference [1] (prior work in the same research line) without re-derivation, external benchmark, or verification that the control enters compatibly with the microscale heterogeneity. This matches the self-citation load-bearing pattern at a moderate level; the central claim retains independent mathematical content beyond the citation.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claim rests on the validity of the heterogeneous SI macroscopic reduction from reference [1] and on the applicability of the Pontryagin Minimum Principle to the resulting constrained optimization problem; no free parameters, invented entities, or additional ad-hoc axioms are stated in the abstract.

axioms (2)
  • domain assumption The macroscopic counterpart of the heterogeneous SI dynamics from reference [1] accurately represents population-level heterogeneity in resistance and viral load.
    The optimal control problem is explicitly constrained to this reduced dynamics.
  • standard math The Pontryagin Minimum Principle applies directly to the resulting optimal control problem with the integral equality constraint.
    The abstract states that PMP is used to obtain the characterization.

pith-pipeline@v0.9.1-grok · 5637 in / 1349 out tokens · 27945 ms · 2026-07-02T02:00:11.455727+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

47 extracted references · 47 canonical work pages · 1 internal anchor

  1. [1]

    Lorenzi, E

    T. Lorenzi, E. Paparelli, A. Tosin,Modelling coevolutionary dynamics in heterogeneous SI epi- demiological systems across scales, Communications in Mathematical Sciences 22 (2024), no. 8, 2131–2165

    work page 2024

  2. [2]

    Fleming, R

    W. Fleming, R. Rishel,Deterministic and Stochastic Optimal Control, Springer-Verlag, 1975

    work page 1975

  3. [3]

    R. F. Hartl, S. P. Sethi, R. G. Vickson,A survey of the Maximum Principles for Optimal Control Problems with State Constraint, SIAM Review 37 (1995), no. 2, 181–218

    work page 1995

  4. [4]

    S. Lee, R. Morales, C. Castillo-Chavez,A note on the use of influenza vaccination strategies when supply is limited, Mathematical Biosciences and Engineering 8 (2011), no. 1, 171–182

    work page 2011

  5. [5]

    M. R. de Pinho, I. Kornienko, H. Maurer,Optimal control of a SEIR model with mixed constraints and L1 cost, in: Proceedings of the 11th Portuguese Conference on Automatic Control, Lecture Notes in Electrical Engineering, vol. 321, Springer, 2015, 135–145

    work page 2015

  6. [6]

    M. H. A. Biswas, L. T. Paiva, M. R. de Pinho,A SEIR model for control of infectious diseases with constraints, Mathematical Biosciences and Engineering 11 (2014), no. 4, 761–784

    work page 2014

  7. [7]

    Optimizing vaccine allocation in an age-structured SIR model

    L. Almeida, R. Ducasse, E. Paparelli, Optimizing vaccine allocation in an age-structured SIR model, arXiv:2605.18256, 2026, https://arxiv.org/abs/2605.18256. 20

    work page internal anchor Pith review Pith/arXiv arXiv 2026

  8. [8]

    Fourer, D

    R. Fourer, D. M. Gay, B. W. Kernighan,AMPL: A Modeling Language for Mathematical Pro- gramming, Duxbury Press, 1993

    work page 1993

  9. [9]

    Wächter, L

    A. Wächter, L. T. Biegler,On the implementation of an interior-point filter line-search algorithm for large-scale nonlinear programming, Math. Program. 106 (2006), 25–57

    work page 2006

  10. [10]

    Behncke,Optimal control of deterministic epidemics, Optimal Control Applications and Meth- ods 21 (2000), no

    H. Behncke,Optimal control of deterministic epidemics, Optimal Control Applications and Meth- ods 21 (2000), no. 6, 269–285

    work page 2000

  11. [11]

    Cianfanelli, F

    L. Cianfanelli, F. Parise, D. Acemoglu, G. Como, A. Ozdaglar,Lockdown interventions in SIR models: Is the reproduction number the right control variable?, in: 2021 60th IEEE Conference on Decision and Control (CDC), 2021, 4254–4259

    work page 2021

  12. [12]

    R. M. Neilan, S. M. Lenhart,An Introduction to Optimal Control with an Application in Disease Modeling, in: Modeling Paradigms and Analysis of Disease Transmission Models, 2010

    work page 2010

  13. [13]

    S. P. Sethi, P. W. Staats,Optimal Control of Some Simple Deterministic Epidemic Models, The Journal of the Operational Research Society 29 (1978), no. 2, 129–136

    work page 1978

  14. [14]

    Sharomi, T

    O. Sharomi, T. Malik,Optimal control in epidemiology, Annals of Operations Research 251 (2017), 55–71

    work page 2017

  15. [15]

    F. Lin, K. Muthuraman, M. Lawley,An optimal control theory approach to non-pharmaceutical interventions, BMC Infectious Diseases 10 (2010), no. 32

    work page 2010

  16. [16]

    Kruse, P

    T. Kruse, P. Strack,Optimal control of an epidemic through social distancing, SSRN Electronic Journal 3581295 (2020)

    work page 2020

  17. [17]

    Ledzewicz, H

    U. Ledzewicz, H. Schättler,On optimal singular controls for a general SIR-model with vaccination and treatment, Conference Publications 2011 (2011), no. Special, 981–990

    work page 2011

  18. [18]

    Bolzoni, E

    L. Bolzoni, E. Bonacini, C. Soresina, M. Groppi,Time-optimal control strategies in SIR epidemic models, Mathematical Biosciences 292 (2017), 86–96

    work page 2017

  19. [19]

    Bolzoni, E

    L. Bolzoni, E. Bonacini, R. Della Marca, M. Groppi,Optimal control of epidemic size and duration with limited resources, Mathematical Biosciences 315 (2019), 108232

    work page 2019

  20. [20]

    Flaxman, S

    S. Flaxman, S. Mishra, A. Gandy, H. J. T. Unwin, T. A. Mellan, H. Coupland, C. Whittaker, Estimating the effects of non-pharmaceutical interventions on COVID-19 in Europe, Nature 584 (2020), 257–261

    work page 2020

  21. [21]

    Gatto, E

    M. Gatto, E. Bertuzzo, L. Mari, S. Miccoli, L. Carraro, R. Casagrandi, A. Rinaldo,Spread and dynamics of the COVID-19 epidemic in Italy: Effects of emergency containment measures, Pro- ceedings of the National Academy of Sciences 117 (2020), no. 19, 10484–10491

    work page 2020

  22. [22]

    A. J. Kucharski, T. W. Russell, C. Diamond, Y. Liu, J. W. Edmunds, S. Funk, R. M. Eggo,Early dynamics of transmission and control of COVID-19: a mathematical modelling study, The Lancet Infectious Diseases 20 (2020), no. 5, 553–558. 21

    work page 2020

  23. [23]

    Giordano, M

    G. Giordano, M. Colaneri, A. Di Filippo, F. Blanchini, P. Bolzern, G. De Nicolao, P. Sacchi, P. Colaneri, R. Bruno,Modeling vaccination rollouts, SARS-CoV-2 variants and the requirement for non-pharmaceutical interventions in Italy, Nature Medicine 27 (2021), no. 6, 993–998

    work page 2021

  24. [24]

    Acemoglu, V

    D. Acemoglu, V. Chernozhukov, I. Werning, M. D. Whinston,Optimal Targeted Lockdowns in a Multigroup SIR Model, American Economic Review: Insights 3 (2021), no. 4, 487–502

    work page 2021

  25. [25]

    J. A. M. Gondim, L. Machado,Optimal quarantine strategies for the COVID-19 pandemic in a population with a discrete age structure, Chaos, Solitons and Fractals 140 (2020), 110166

    work page 2020

  26. [26]

    X. Lü, H. Hui, F. Liu, Stability and optimal control strategies for a novel epidemic model of COVID-19, Nonlinear Dynamics 106 (2021), 1491–1507

    work page 2021

  27. [27]

    Mandal, S

    M. Mandal, S. Jana, S. K. Nandi, A. Khatua, S. Adak, T. K. Kar,A model based study on the dynamics of COVID-19: Prediction and control, Chaos, Solitons and Fractals 136 (2020), 109889

    work page 2020

  28. [28]

    G. B. Libotte, F. S. Lobato, G. M. Platt, A. J. Silva Neto,Determination of an optimal control strategy for vaccine administration in COVID-19 pandemic treatment, Computer Methods and Programs in Biomedicine 196 (2020), 105664

    work page 2020

  29. [29]

    Abbasi, I

    Z. Abbasi, I. Zamani, A. H. A. Mehra, M. Shafieirad, A. Ibeas,Optimal Control Design of Impul- sive SQEIAR Epidemic Models with Application to COVID-19, Chaos, Solitons and Fractals 139 (2020), 110054

    work page 2020

  30. [30]

    J. K. K. Asamoah, Z. Jin, G.-Q. Sun, B. Seidu, E. Yankson, A. Abidemi, F. T. Oduro, S. E. Moore, E. Okyere,Sensitivity assessment and optimal economic evaluation of a new COVID-19 compartmental epidemic model with control interventions, Chaos, Solitons and Fractals 146 (2021), 110885

    work page 2021

  31. [31]

    W. Choi, E. Shim, Optimal strategies for social distancing and testing to control COVID-19, Journal of Theoretical Biology 512 (2021), 110568

    work page 2021

  32. [32]

    Mondal, S

    J. Mondal, S. Khajanchi, Mathematical modeling and optimal intervention strategies of the COVID-19 outbreak, Nonlinear Dynamics 109 (2022), 177–202

    work page 2022

  33. [33]

    Shen, Y.-M

    Z.-H. Shen, Y.-M. Chu, M. A. Khan, S. Muhammad, O. A. Al-Hartomy, M. Higazy,Mathematical modeling and optimal control of the COVID-19 dynamics, Results in Physics 31 (2021), 105028

    work page 2021

  34. [34]

    J.Köhler, L.Schwenkel, A.Koch, J.Berberich, P.Pauli, F.Allgöwer,Robust and optimal predictive control of the COVID-19 outbreak, Annual Reviews in Control 51 (2021), 525–539

    work page 2021

  35. [35]

    Miclo, D

    L. Miclo, D. Spiro, J. Weibull,Optimal epidemic suppression under an ICU constraint: An ana- lytical solution, Journal of Mathematical Economics 101 (2022), 102669

    work page 2022

  36. [36]

    Paparelli, R

    E. Paparelli, R. Giambó, H. Maurer,Optimal control of an epidemiological Covid-19 model with state constraint, Discrete and Continuous Dynamical Systems - B 30 (2025), no. 2, 422–448. 22

    work page 2025

  37. [37]

    M. R. De Pinho, I. Kornienko, H. Maurer,Optimal control of a SEIR model with mixed constraints and L1 cost, Lecture Notes in Electrical Engineering 321 (2015), 135–145

    work page 2015

  38. [38]

    M. H. A. Biswas, L. T. Paiva, M. R. De Pinho,A SEIR model for control of infectious diseases with constraints, Mathematical Biosciences and Engineering 11 (2014), no. 4, 761–784

    work page 2014

  39. [39]

    A.Charpentier, R.Elie, M.Laurière, V.C.Tran, COVID-19 pandemic control: balancing detection policy and lockdown intervention under ICU sustainability, Math. Model. Nat. Phenom. 15 (2020), 57

    work page 2020

  40. [40]

    J. P. Caulkins, D. Grass, G. Feichtinger, R. F. Hartl, P. K. Kort, A. Prskawetz, A. Seidl, S. Wrzaczek,The optimal lockdown intensity for COVID-19, Journal of Mathematical Economics 93 (2021), 102489

    work page 2021

  41. [41]

    flattening the curve

    M. Kantner, T. Koprucki,Beyond just "flattening the curve": Optimal control of epidemics with purely non-pharmaceutical interventions, Journal of Mathematics in Industry 10 (2020), no. 1

    work page 2020

  42. [42]

    Aspri, E

    A. Aspri, E. Beretta, A. Gandolfi, E. Wasmer,Mortality containment vs. Economics Opening: Optimal policies in a SEIARD model, Journal of Mathematical Economics 93 (2021)

    work page 2021

  43. [43]

    Franceschi, A

    J. Franceschi, A. Medaglia, M. Zanella,On the optimal control of kinetic epidemic models with uncertain social features, Optimal Control, Applications and Methods 45 (2024), no. 2, 494–522

    work page 2024

  44. [44]

    Dimarco, G

    G. Dimarco, G. Toscani, M. Zanella,Optimal control of epidemic spreading in the presence of social heterogeneity, Philosophical Transactions of the Royal Society A 380 (2022), 20210160

    work page 2022

  45. [45]

    G. Albi, L. Pareschi, M. Zanella,Control with uncertain data of socially structured compartmental epidemic models, Journal of Mathematical Biology 82 (2021), 63

    work page 2021

  46. [46]

    G. Albi, L. Pareschi, M. Zanella,Modelling lockdown measures in epidemic outbreaks using selec- tive socio-economic containment with uncertainty, Mathematical Biosciences and Engineering 18 (2021), no. 6, 7161–7190

    work page 2021

  47. [47]

    https://neos-server.org/neos/index.html . 23

    work page