Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem 2, p. 216 — the limit SDE dq=f(q) dt+2μ dbdq = f(q)\,dt + \sqrt{2\mu}\,dbdq=f(q)dt+2μ​db has a unique strong solution

Proved
ErlangA.Diffusion.sde_well_posed

by mikedeng1 · Oct 7, 2026 · Mathlib 0df444a (Lean v4.33.1)

diffusion-approximationp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1stochastic-differential-equations

Let μ>0\mu > 0μ>0, θ>0\theta > 0θ>0 and β∈R\beta \in \mathbb Rβ∈R, and let

f(x)={−μ(β+x),x≤0,−(μβ+θx),x>0.f(x) = \begin{cases} -\mu(\beta + x), & x \le 0,\\ -(\mu\beta + \theta x), & x > 0.\end{cases}f(x)={−μ(β+x),−(μβ+θx),​x≤0,x>0.​

Let (Ω,F,P)(\Omega, \mathcal F, P)(Ω,F,P) be a probability space carrying a standard Brownian motion bbb (every path continuous) and a real random variable X0X_0X0​ independent of bbb. Then:

  1. Existence. There is a process qqq with q(0)=X0q(0) = X_0q(0)=X0​, continuous paths on [0,∞)[0,\infty)[0,∞), each q(t)q(t)q(t) measurable with respect to σ(X0,b(s):s≤t)\sigma(X_0, b(s) : s \le t)σ(X0​,b(s):s≤t), such that almost surely, for all t≥0t \ge 0t≥0,
q(t)=q(0)+∫0tf(q(s)) ds+2μ b(t).q(t) = q(0) + \int_0^t f(q(s))\,ds + \sqrt{2\mu}\, b(t).q(t)=q(0)+∫0t​f(q(s))ds+2μ​b(t).
  1. Pathwise uniqueness. If qqq and q′q'q′ both solve this equation driven by the same bbb and q(0)=q′(0)q(0) = q'(0)q(0)=q′(0) almost surely, then almost surely q(t)=q′(t)q(t) = q'(t)q(t)=q′(t) for all t≥0t \ge 0t≥0.

This makes "qqq is the unique solution of the stochastic differential equation" in Theorem 2 meaningful; the paper omits the discussion and refers to Dupuis and Ishii (1993) and Mandelbaum and Pats (1995).

Formalization Note Because the noise is additive and fff is Lipschitz, the equation is written pathwise with a Lebesgue integral; a solution is a strong solution. The hypotheses θ>0\theta > 0θ>0, μ>0\mu > 0μ>0 are those of Theorem 2.

Preamble
import Mathlib
import Definitions.Def_ErlangA_Diffusion_SDE

open MeasureTheory ProbabilityTheory Filter Topology
open scoped ENNReal NNReal
Formal statement
namespace ErlangA.Diffusion

/-- Well-posedness of the limit SDE of Theorem 2 (p. 216; p. 224). Let `μ > 0`, `θ > 0` and
`β ∈ ℝ`. On a probability space carrying a standard Brownian motion `b` (every path continuous)
and a real random variable `X₀` independent of `b`:
1. (existence) there is a process `q` with `q(0) = X₀`, adapted to the filtration generated by
   `X₀` and `b`, solving `dq(t) = f(q) dt + √(2μ) db(t)`;
2. (pathwise uniqueness) any two solutions driven by the same `b` with almost surely equal
   initial values are indistinguishable on `[0, ∞)`. -/
theorem sde_well_posed (μ β θ : ℝ) (hμ : 0 < μ) (hθ : 0 < θ)
    {Ω : Type*} [MeasurableSpace Ω] (P : Measure Ω) [IsProbabilityMeasure P]
    (b : ℝ≥0 → Ω → ℝ) (hb : IsStandardBM P b)
    (X0 : Ω → ℝ) (hX0 : Measurable X0)
    (hind : IndepFun X0 (fun ω (t : ℝ≥0) => b t ω) P) :
    (∃ q : Ω → ℝ → ℝ, (∀ ω, q ω 0 = X0 ω) ∧ IsAdaptedToInitialAndBM b X0 q ∧
      SolvesDiffusionSDE P μ β θ b q) ∧
    ∀ q q' : Ω → ℝ → ℝ, SolvesDiffusionSDE P μ β θ b q → SolvesDiffusionSDE P μ β θ b q' →
      (∀ᵐ ω ∂P, q ω 0 = q' ω 0) → ∀ᵐ ω ∂P, ∀ t : ℝ, 0 ≤ t → q ω t = q' ω t := by sorry

end ErlangA.Diffusion
Source
Garnett, Mandelbaum & Reiman, Designing a Call Center with Impatient Customers, M&SOM 4(3), 2002, p. 216, Theorem 2 ("q is the unique solution"); p. 224, App. C (uniqueness remark)
Human review
  • Endorsed by Shuze Chen · Oct 7, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 7, 2026

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

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