pith. sign in

arxiv: 2606.17217 · v1 · pith:UCJ6XPBInew · submitted 2026-06-15 · 📡 eess.SY · cs.SY

A Stateful Stochastic Allocation Mechanism with Fairness Guarantees for Networked Electricity Systems

Pith reviewed 2026-06-27 02:49 UTC · model grok-4.3

classification 📡 eess.SY cs.SY
keywords fairness guaranteesstochastic allocationelectricity marketsDC-OPFLyapunov analysisconvergence boundsstateful mechanismautomatic market maker
0
0 comments X

The pith

The FP-AMM ensures per-node delivery ratios converge almost surely to contracted fairness targets F^* with O(1/sqrt(T)) bounds.

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

This paper introduces the Fair Play Automatic Market Maker, a stateful mechanism for allocating scarce electricity that remembers past shortages to enforce fairness over time. Existing locational marginal pricing approaches are memoryless and fail to guarantee equitable service across intervals. The mechanism uses a two-stage stochastic clearing process and proves almost sure convergence of delivery ratios to targets, along with contraction properties and bounded tracking error under event triggering. Simulations on IEEE bus systems confirm reduced fairness errors compared to baselines while maintaining power flow feasibility.

Core claim

The Fair Play Automatic Market Maker employs a two-stage stochastic clearing rule with service-priority sampling and inverse-fairness weighting, coupled with a DC-OPF feasibility set and a saturated integrator for shortage memory. Under the Fair Play priority rule, the per-node delivery ratio converges almost surely to the contracted target F^*, with a finite-time O(1/sqrt(T)) bound obtained via Lyapunov analysis of the deficit recursion. The shortage-memory state is invariant in [0,1]^N with contraction rate 1-β, and the intra-interval clearing operator converges linearly to a unique fixed point.

What carries the argument

The deficit recursion updated by a saturated integrator under the Fair Play priority rule, analyzed via Lyapunov methods to establish almost-sure convergence.

If this is right

  • The intra-interval clearing operator converges linearly to a unique fixed point with contraction factor q in (0,1).
  • The shortage-memory state is invariant in [0,1]^N and the update map is a contraction with rate 1-β.
  • Event-triggered execution guarantees practical ultimate boundedness of the allocation tracking error.
  • Fairness convergence to F^* is achieved on all IEEE benchmark networks with peak weak-bus fairness error reduced by up to 55% during scarcity.
  • DC feasibility is maintained throughout the market intervals.

Where Pith is reading between the lines

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

  • The stateful approach could apply to other scarcity allocation problems in infrastructure networks where memory of past service is needed for equity.
  • Contraction mapping properties indicate the mechanism may not require topology-specific adjustments for stability.
  • Event-triggered execution provides a tunable trade-off between computation and fairness performance in operational settings.
  • Integration with variable renewable sources could leverage the boundedness guarantees to handle intermittent supply while preserving fairness.

Load-bearing premise

The two-stage stochastic clearing rule combined with the DC-OPF feasibility set produces a well-defined fixed point whose stability is independent of the specific network topology and load realizations.

What would settle it

Running the mechanism on a network for large T and observing that the per-node delivery ratio does not approach F^* or that the O(1/sqrt(T)) bound is violated would falsify the almost-sure convergence claim.

Figures

Figures reproduced from arXiv: 2606.17217 by Shaun Sweeney.

Figure 1
Figure 1. Figure 1: shows [PITH_FULL_IMAGE:figures/full_fig_p007_1.png] view at source ↗
Figure 4
Figure 4. Figure 4: shows how the trigger threshold δ affects update frequency and tracking error. Increasing δ reduces the fraction of intervals with updates, at the cost of larger mean and 95th￾percentile tracking error, consistent with the practical ultimate bound (21) from Theorem 4(c). The dashed line shows the predicted bound using q = 0.946, K = 20, and L⋆ = 18.5 from Table I. 2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0 Trigg… view at source ↗
Figure 5
Figure 5. Figure 5: Fairness as bounded perturbation: allocation changes without destabilisation [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
Figure 7
Figure 7. Figure 7: Maximum fairness error on IEEE benchmarks (Fair Play ON). maxn |1 − Fn(t)| converges to zero on all three networks after the scarcity windows, consistent with Theorem 5. IEEE-57 converges fastest, consistent with its largest estimated ch (Table II). 0 1000 2000 3000 4000 5000 Market interval 0.75 0.80 0.85 0.90 0.95 1.00 Fairn ess ratio Fn(t) Fairness convergence ieee_57_fair_on Weak (remote) buses Strong … view at source ↗
Figure 8
Figure 8. Figure 8: Per-bus fairness convergence, IEEE-57-bus (Fair Play ON). Red: weak (remote) buses; blue: strong (proximate) buses. All 57 buses converge to F ⋆ = 1 despite the persistent 1.8× demand disadvantage on weak buses, directly illustrating Theorem 5(a). B. Fairness Convergence on IEEE Benchmarks [PITH_FULL_IMAGE:figures/full_fig_p009_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Fair Play ON vs OFF: weak-bus correction across all benchmarks. Top row: maxn |1 − Fn(t)|; bottom row: mean weak-bus error |1 − Fn(t)|. Columns left to right: IEEE-14, IEEE-57, IEEE-118. Solid blue: Fair Play ON; dashed red: Fair Play OFF. During scarcity windows (t ≈ 400–900 and 1600–2200), Fair Play ON maintains materially lower peak error on all three networks. The separation is strongest on IEEE-57, wh… view at source ↗
Figure 10
Figure 10. Figure 10: DC feasibility: maximum network loading (Fair Play ON). Maximum line/transformer loading on all three benchmarks remains well below the 70% feasibility limit throughout the T = 5000 simulation. [10] W. P. M. H. Heemels, M. C. F. Donkers, and A. R. Teel, “Periodic event-triggered control for linear systems,” IEEE Trans. Autom. Control, vol. 58, no. 4, pp. 847–861, 2013. [11] D. P. Bertsekas and R. G. Galla… view at source ↗
Figure 11
Figure 11. Figure 11: Event-trigger update rate vs network size. Fraction of intervals in which the clearing operator is re-solved. The rate varies non-monotonically with network size, reflecting network-specific state-change dynamics, and never reaches 100%, confirming the event-triggered reduction in computation consistent with Theorem 4 [PITH_FULL_IMAGE:figures/full_fig_p011_11.png] view at source ↗
read the original abstract

This paper develops and analyses the Fair Play Automatic Market Maker (FP-AMM), a programmable electricity allocation mechanism in which scarcity allocation is treated as a controlled, stateful, and auditable cyber-physical process. Existing mechanisms such as locational marginal pricing are memoryless and cannot account for historical service outcomes, preventing guarantees of equitable treatment across market intervals. The FP-AMM employs a two-stage stochastic clearing rule comprising service-priority sampling and inverse-fairness weighting, coupled with a DC-OPF feasibility set and bounded shortage memory updated through a saturated integrator. Four main results are established. First, the shortage-memory state is invariant in $[0,1]^N$ and the update map is a contraction with rate $1-\beta$. Second, the intra-interval clearing operator converges linearly to a unique fixed point with contraction factor $q\in(0,1)$. Third, under the Fair Play priority rule, the per-node delivery ratio converges almost surely to the contracted target $F^\star$, with a finite-time $O(1/\sqrt{T})$ bound obtained via Lyapunov analysis of the deficit recursion. Fourth, event-triggered execution guarantees practical ultimate boundedness of the allocation tracking error and quantifies the computation-fidelity trade-off. The mechanism is validated on the IEEE 14-, 57-, and 118-bus systems over $T=5000$ market intervals. Fairness convergence to $F^\star$ is achieved on all benchmarks, peak weak-bus fairness error is reduced by 54% on the IEEE-57 network and by up to 55% relative to an equal-weight baseline during scarcity periods, and DC feasibility is maintained throughout.

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

3 major / 2 minor

Summary. The paper proposes the Fair Play Automatic Market Maker (FP-AMM), a stateful stochastic allocation mechanism for networked electricity systems. It uses a two-stage clearing rule (service-priority sampling plus inverse-fairness weighting) projected onto a DC-OPF feasible set, together with a bounded shortage-memory state updated by a saturated integrator. Four convergence results are claimed: invariance of the memory state in [0,1]^N with contraction rate 1-β; linear convergence of the intra-interval clearing operator to a unique fixed point with rate q∈(0,1); almost-sure convergence of per-node delivery ratios to a target F* with an O(1/√T) finite-time bound via Lyapunov analysis; and practical ultimate boundedness under event-triggered execution. The claims are supported by numerical experiments on the IEEE 14-, 57-, and 118-bus systems over T=5000 intervals, showing fairness improvements relative to an equal-weight baseline.

Significance. If the contraction factor q can be shown to be uniformly bounded away from 1 independently of network topology, line limits, and load realizations, the work would provide a concrete, auditable mechanism that delivers explicit fairness guarantees across market intervals, addressing a recognized limitation of memoryless locational marginal pricing. The combination of Lyapunov-based finite-time bounds, event-triggered implementation, and validation on standard test systems would constitute a substantive contribution to stochastic resource allocation in cyber-physical energy systems.

major comments (3)
  1. [§4.2] §4.2 (clearing operator convergence): The proof that the composite map (two-stage rule projected onto the DC-OPF polytope) is a contraction with factor q<1 independent of topology must explicitly bound the Lipschitz constant of the DC-OPF projection. Because the feasible set geometry changes with line limits and realized loads, it is not immediate that q remains uniformly bounded away from 1; this bound is load-bearing for the subsequent a.s. convergence and O(1/√T) claim in Theorem 3.
  2. [Theorem 3] Theorem 3 (Lyapunov analysis of deficit recursion): The finite-time O(1/√T) bound and almost-sure convergence to F* treat the per-interval allocation as being within O(q^t) of the fixed point. If the contraction rate q can approach 1 for admissible networks or load patterns (as permitted by the DC-OPF set), the uniform bound fails; the manuscript must either derive a topology-independent upper bound on q or state the precise conditions under which the result holds.
  3. [§3] Definition of F* and the target contraction: The abstract and §3 describe convergence to the 'contracted target F*'. It must be clarified whether F* is an exogenous design parameter or an emergent quantity of the fixed-point map; if the latter, the proof must show that the map's fixed point coincides with the chosen F* without circularity.
minor comments (2)
  1. [Numerical experiments] The numerical section should report the empirical distribution of the observed contraction rates across the 5000 intervals and the three test systems, together with the maximum observed q, to allow readers to assess how close q remains to 1 in practice.
  2. [§3.1] Notation for the saturated integrator and the inverse-fairness weights should be introduced with explicit equations before the theorems that rely on them.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the thorough review and insightful comments on the contraction properties and the role of F*. We address each major comment below, indicating revisions where the manuscript requires clarification or additional statements of conditions.

read point-by-point responses
  1. Referee: [§4.2] §4.2 (clearing operator convergence): The proof that the composite map (two-stage rule projected onto the DC-OPF polytope) is a contraction with factor q<1 independent of topology must explicitly bound the Lipschitz constant of the DC-OPF projection. Because the feasible set geometry changes with line limits and realized loads, it is not immediate that q remains uniformly bounded away from 1; this bound is load-bearing for the subsequent a.s. convergence and O(1/√T) claim in Theorem 3.

    Authors: The proof in §4.2 establishes that the composite clearing operator is a contraction mapping with factor q ∈ (0,1) for any fixed network topology, line limits, and load realization, by showing that the two-stage stochastic rule followed by Euclidean projection onto the DC-OPF polytope yields a unique fixed point. However, the Lipschitz constant of the projection operator does depend on the geometry of the feasible set and is not claimed to be uniformly bounded away from 1 independently of topology or loads. We will revise §4.2 to explicitly state this dependence and add a sufficient condition (strict feasibility of the DC-OPF set with positive margin on all lines) under which q can be bounded by a constant strictly less than 1 for a given network class. This preserves the per-instance convergence result while clarifying the scope. revision: partial

  2. Referee: [Theorem 3] Theorem 3 (Lyapunov analysis of deficit recursion): The finite-time O(1/√T) bound and almost-sure convergence to F* treat the per-interval allocation as being within O(q^t) of the fixed point. If the contraction rate q can approach 1 for admissible networks or load patterns (as permitted by the DC-OPF set), the uniform bound fails; the manuscript must either derive a topology-independent upper bound on q or state the precise conditions under which the result holds.

    Authors: Theorem 3 derives the O(1/√T) finite-time bound and a.s. convergence under the assumption that the intra-interval operator has already reached a neighborhood of its fixed point (controlled by the linear rate q). Because q is instance-dependent, the bound holds for each fixed network but is not uniform across all admissible DC-OPF instances. We will revise the statement of Theorem 3 and its proof to include an explicit remark that the result applies whenever q ≤ q_max < 1 for the given network parameters, and we will add a corollary quantifying the degradation of the constant when q approaches 1. No topology-independent bound is derived, as the referee correctly notes this would require additional assumptions on line capacities and load statistics. revision: partial

  3. Referee: [§3] Definition of F* and the target contraction: The abstract and §3 describe convergence to the 'contracted target F*'. It must be clarified whether F* is an exogenous design parameter or an emergent quantity of the fixed-point map; if the latter, the proof must show that the map's fixed point coincides with the chosen F* without circularity.

    Authors: F* is an exogenous design parameter chosen by the system operator to encode the desired long-term fairness level (a vector in (0,1)^N). The two-stage clearing rule (service-priority sampling followed by inverse-fairness weighting) is deliberately constructed so that its unique fixed point, after projection, produces per-node delivery ratios exactly equal to F*. The proof in §3 first defines the target F*, then verifies that the fixed-point equation of the composite operator is satisfied precisely when the delivery ratios equal F*, without circular reasoning. We will revise the abstract and §3 to state this explicitly and add a short lemma confirming the fixed-point coincidence. revision: yes

Circularity Check

0 steps flagged

No circularity; convergence results derived from explicit recursions and Lyapunov analysis

full rationale

The paper defines the FP-AMM mechanism, its two-stage clearing rule, shortage memory update, and Fair Play priority explicitly, then applies standard contraction mapping and Lyapunov arguments to prove invariance, linear convergence of the clearing operator, and almost-sure tracking of the design parameter F*. These steps rely on the stated assumptions about the DC-OPF set and bounded integrator rather than reducing the claimed outcomes to the inputs by definition or via self-citation. No load-bearing step equates a derived quantity to a fitted or renamed input; the finite-time bound and a.s. convergence are obtained from the deficit recursion analysis, which is independent of the target value itself.

Axiom & Free-Parameter Ledger

3 free parameters · 2 axioms · 0 invented entities

The central claims rest on standard contraction-mapping and Lyapunov arguments plus the modeling choice that shortage memory can be bounded in [0,1]^N without violating DC-OPF feasibility. No new physical entities are postulated.

free parameters (3)
  • beta
    Contraction rate of the shortage-memory update map; appears as a tunable gain in the saturated integrator.
  • F*
    Target fairness level to which delivery ratios converge; described as 'contracted' and therefore chosen by the designer.
  • q
    Contraction factor of the intra-interval clearing operator; stated to lie in (0,1) but its explicit dependence on network parameters is not given in the abstract.
axioms (2)
  • domain assumption The DC-OPF feasibility set remains non-empty under the stochastic priority sampling rule.
    Invoked to guarantee that every market clearing produces a feasible dispatch.
  • standard math The shortage-memory update is a contraction mapping with rate 1-beta on the compact set [0,1]^N.
    Used to establish invariance and convergence of the memory state.

pith-pipeline@v0.9.1-grok · 5827 in / 1573 out tokens · 56940 ms · 2026-06-27T02:49:12.388077+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

18 extracted references

  1. [1]

    Programmable fairness in electricity markets: A cost- causation–consistent alternative to marginal pricing,

    S. Sweeney, “Programmable fairness in electricity markets: A cost- causation–consistent alternative to marginal pricing,”Energy Economics, vol. 158, p. 109345, 2026

  2. [2]

    F. C. Schweppe, M. C. Caramanis, R. D. Tabors, and R. E. Bohn,Spot Pricing of Electricity. Kluwer Academic Publishers, 1988

  3. [3]

    Stoft,Power System Economics: Designing Markets for Electricity

    S. Stoft,Power System Economics: Designing Markets for Electricity. IEEE Press/Wiley, 2002

  4. [4]

    A. J. Wood, B. F. Wollenberg, and G. B. Shebl ´e,Power Generation, Operation, and Control, 3rd ed. Wiley, 2013

  5. [5]

    A. R. Bergen and V . Vittal,Power Systems Analysis, 2nd ed. Prentice Hall, 2000

  6. [6]

    Optimization flow control, I: Basic algorithm and convergence,

    S. H. Low and D. E. Lapsley, “Optimization flow control, I: Basic algorithm and convergence,”IEEE/ACM Trans. Netw., vol. 7, no. 6, pp. 861–874, 1999

  7. [7]

    A tutorial on decomposition methods for network utility maximization,

    D. P. Palomar and M. Chiang, “A tutorial on decomposition methods for network utility maximization,”IEEE J. Sel. Areas Commun., vol. 24, no. 8, pp. 1439–1451, 2006

  8. [8]

    A generalized processor sharing approach to flow control in integrated services networks,

    A. K. Parekh and R. G. Gallager, “A generalized processor sharing approach to flow control in integrated services networks,”IEEE/ACM Trans. Netw., vol. 1, no. 3, pp. 344–357, 1993

  9. [9]

    Event-triggered real-time scheduling of stabilizing control tasks,

    P. Tabuada, “Event-triggered real-time scheduling of stabilizing control tasks,”IEEE Trans. Autom. Control, vol. 52, no. 9, pp. 1680–1685, 2007. 11 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 maxn |1 Fn (t)| IEEE-14 IEEE-57 IEEE-118 Fair Play ON Fair Play OFF 0 1000 2000 3000 4000 5000 Market interval 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 Mean |1 Fn(t)| (weak buses) 0 1000...

  10. [10]

    Periodic event-triggered control for linear systems,

    W. P. M. H. Heemels, M. C. F. Donkers, and A. R. Teel, “Periodic event-triggered control for linear systems,”IEEE Trans. Autom. Control, vol. 58, no. 4, pp. 847–861, 2013

  11. [11]

    D. P. Bertsekas and R. G. Gallager,Data Networks, 2nd ed. Prentice Hall, 1992

  12. [12]

    Rate control for commu- nication networks: shadow prices, proportional fairness and stability,

    F. P. Kelly, A. K. Maulloo, and D. K. H. Tan, “Rate control for commu- nication networks: shadow prices, proportional fairness and stability,”J. Oper. Res. Soc., vol. 49, no. 3, pp. 237–252, 1998

  13. [13]

    Efficient fair queuing using deficit round robin,

    M. Shreedhar and G. Varghese, “Efficient fair queuing using deficit round robin,”IEEE/ACM Trans. Netw., vol. 4, no. 3, pp. 375–385, 1996

  14. [14]

    A stochastic approximation method,

    H. Robbins and S. Monro, “A stochastic approximation method,”Ann. Math. Statist., vol. 22, no. 3, pp. 400–407, 1951

  15. [15]

    V . S. Borkar,Stochastic Approximation: A Dynamical Systems View- point. Cambridge University Press, 2008

  16. [16]

    J. F. Bonnans and A. Shapiro,Perturbation Analysis of Optimization Problems. Springer, 2000

  17. [17]

    A convergence theorem for non negative almost supermartingales and some applications,

    H. Robbins and D. Siegmund, “A convergence theorem for non negative almost supermartingales and some applications,” inOptimizing Methods in Statistics, pp. 233–257, 1971

  18. [18]

    pandapower — an open-source Python tool for con- venient modeling, analysis, and optimization of electric power systems,

    L. Thurneret al., “pandapower — an open-source Python tool for con- venient modeling, analysis, and optimization of electric power systems,” IEEE Trans. Power Syst., vol. 33, no. 6, pp. 6510–6521, 2018. 20 40 60 80 100 120 Number of buses 0.45 0.50 0.55 0.60 0.65 0.70Fraction of intervals updated Event-trigger update rate vs network size Fig. 11.Event-tri...