Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem 2.8 — the product form of an open migration process

Proved
KellyStochasticNetworks.open_migration_equilibrium

by naimengye · Sep 18, 2026 · Mathlib 0df444a (Lean v4.33.1)

markov-chainproduct-formqueueing-theorystochastic-networks

Consider an open migration process on JJJ colonies. The state is the vector n=(n1,…,nJ)n = (n_1,\dots,n_J)n=(n1​,…,nJ​) of colony occupancies, and the transition rates are

q(n,Tjkn)=λjkφj(nj),q(n,Tj→n)=μjφj(nj),q(n,T→kn)=νk,q(n, T^{jk}n) = \lambda_{jk}\varphi_j(n_j), \qquad q(n, T^{j\to}n) = \mu_j\varphi_j(n_j), \qquad q(n, T^{\to k}n) = \nu_k,q(n,Tjkn)=λjk​φj​(nj​),q(n,Tj→n)=μj​φj​(nj​),q(n,T→kn)=νk​,

where TjkT^{jk}Tjk moves one individual from colony jjj to colony kkk, Tj→T^{j\to}Tj→ removes one from colony jjj, and T→kT^{\to k}T→k adds one to colony kkk. All rates are non-negative, λjj=0\lambda_{jj} = 0λjj​=0, and φj(0)=0\varphi_j(0) = 0φj​(0)=0 with φj(r)>0\varphi_j(r) > 0φj​(r)>0 for r≥1r \ge 1r≥1, so nothing leaves an empty colony.

Let (αj)(\alpha_j)(αj​) be a positive solution of the traffic equations (2.2),

αj(μj+∑kλjk)=νj+∑kαkλkj,\alpha_j\Bigl(\mu_j + \sum_k \lambda_{jk}\Bigr) = \nu_j + \sum_k \alpha_k \lambda_{kj},αj​(μj​+k∑​λjk​)=νj​+k∑​αk​λkj​,

and suppose each series

gj=∑m=0∞αj m∏r=1mφj(r)g_j = \sum_{m=0}^{\infty}\frac{\alpha_j^{\,m}}{\prod_{r=1}^{m}\varphi_j(r)}gj​=m=0∑∞​∏r=1m​φj​(r)αjm​​

converges. Then

π(n)=∏j=1Jπj(nj),πj(m)=gj−1 αj m∏r=1mφj(r)\pi(n) = \prod_{j=1}^{J}\pi_j(n_j), \qquad \pi_j(m) = g_j^{-1}\,\frac{\alpha_j^{\,m}}{\prod_{r=1}^{m}\varphi_j(r)}π(n)=j=1∏J​πj​(nj​),πj​(m)=gj−1​∏r=1m​φj​(r)αjm​​

is the equilibrium distribution of the process: it satisfies the equilibrium equations, and it sums to 111 over the whole state space Z+J\mathbb{Z}_+^{J}Z+J​.

The conclusion is that in equilibrium, at any fixed time, the occupancies n1(t),…,nJ(t)n_1(t),\dots,n_J(t)n1​(t),…,nJ​(t) are independent, each distributed as if its colony were fed by a Poisson stream — although in most open migration networks they are not independent as processes, and the arrival stream into a colony is not Poisson.

Formalization Note The convergence hypothesis gj<∞g_j < \inftygj​<∞ and the value of gjg_jgj​ are asserted together as a single summability statement, one per colony. Both halves of the conclusion are stated: the equilibrium equations alone are homogeneous and are satisfied by every positive multiple of π\piπ, so the normalization is what pins the constants gjg_jgj​ down. The equilibrium equations are written against the assembled rate matrix, with sums over actual states, so a transition out of an empty colony contributes nothing rather than being read through truncated subtraction.

Preamble
import Mathlib
import Definitions.Def_KellyStochasticNetworks_Balance
import Definitions.Def_KellyStochasticNetworks_Migration
Formal statement
namespace KellyStochasticNetworks

theorem open_migration_equilibrium {J : ℕ} (lam : Fin J → Fin J → ℝ)
    (mu nu α g : Fin J → ℝ) (φ : Fin J → ℕ → ℝ)
    (hlamdiag : ∀ j, lam j j = 0) (hlam : ∀ j k, 0 ≤ lam j k)
    (hmu : ∀ j, 0 ≤ mu j) (hnu : ∀ j, 0 ≤ nu j)
    (hφ0 : ∀ j, φ j 0 = 0) (hφpos : ∀ j r, 1 ≤ r → 0 < φ j r)
    (hα : ∀ j, 0 < α j) (htraffic : OpenTraffic lam mu nu α)
    (hg : ∀ j, HasSum (fun m : ℕ => α j ^ m / phiProd φ j m) (g j)) :
    FullBalance (openMigrationPi α g φ) (openMigrationRates lam mu nu φ)
      ∧ HasSum (openMigrationPi α g φ) 1 := by sorry

end KellyStochasticNetworks
Source
Kelly & Yudovina, Stochastic Networks, CUP 2014, p. 31 (PDF p. 39), Theorem 2.8: 'Define the constants g_j = sum_{n=0}^{infinity} alpha_j^n / prod_{r=1}^{n} phi_j(r), j = 1, 2, ..., J. Theorem 2.8 If g_1, ..., g_J < infinity, then n has the equilibrium distribution pi(n) = prod_{j=1}^{J} pi_j(n_j), pi_j(n_j) = g_j^{-1} alpha_j^{n_j} / prod_{r=1}^{n_j} phi_j(r).' The traffic equations (2.2) are on the same page. sha256 ec271d555059aee58613e5e9a98b8346214b16185c527314f5d94c1ac8b17b6a
Human review
  • Endorsed by Shuze Chen · Sep 25, 2026

    Confirmed by the moderator at approval.

  • Endorsed by naimengye · Sep 25, 2026

    Confirmed by the mission captain (proposal self-audit).

  • Endorsed by mikedeng1 · Oct 5, 2026

    Confirmed by the mission captain (proposal self-audit).

  • Endorsed by Community (Bot) · Oct 22, 2026

    Confirmed by the moderator at approval.

View graph

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions, with reuse governed by our licensing terms.

How Prove2Me worksResearch paperProve2Me Team
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me