Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Claim 1 — under any policy in Z(s), almost surely the state lies in X(s) from some period on, so all states outside X(s) are transient

Proved
LostSalesOldNew.StateReduction.claim_1

by mikedeng1 · Oct 6, 2026 · Mathlib 0df444a (Lean v4.33.1)

lost-salesmarkov-chainsp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1transience

Consider the lost-sales system with lead time L≥1L\ge 1L≥1 and i.i.d. demands d0,d1,…d_0,d_1,\dotsd0​,d1​,… with common law DDD on [0,∞)[0,\infty)[0,∞), where demand is not almost surely zero: P(d>0)>0P(d>0)>0P(d>0)>0. Let s=(s0,…,sL)s=(s_0,\dots,s_L)s=(s0​,…,sL​) satisfy s0≥s1≥⋯≥sL≥0s_0\ge s_1\ge\dots\ge s_L\ge 0s0​≥s1​≥⋯≥sL​≥0, let z∈Z(s)z\in Z(s)z∈Z(s) be a stationary policy, and let x≥0x\ge 0x≥0 be any initial state. Let xtx_txt​ be the state in period ttt under zzz, started at x0=xx_0=xx0​=x. Then, almost surely,

∃ T  ∀ t≥T:xt∈X(s).\exists\, T\ \ \forall\, t\ge T:\qquad x_t\in X(s).∃T  ∀t≥T:xt​∈X(s).

The paper states: "For any policy in Z(s)Z(s)Z(s), all states outside X(s)X(s)X(s) are transient." We state: almost surely the state lies in X(s)X(s)X(s) from some period on, so every state outside X(s)X(s)X(s) (indeed the whole complement of X(s)X(s)X(s)) is visited only finitely often. This is the reading the paper's proof establishes — the process started outside X(s)X(s)X(s) eventually reaches X(s)X(s)X(s), and once it enters X(s)X(s)X(s) it never leaves.

Consequently, to solve the dynamic program one may restrict attention to the compact state space X(sˉ)X(\bar s)X(sˉ) and the policies Z(sˉ)Z(\bar s)Z(sˉ).

Formalization Note. The demand sequence is the coordinate process under the product measure D⊗ND^{\otimes\mathbb N}D⊗N (Measure.infinitePi), which encodes the paper's independent, stationary, nonnegative demands. The hypothesis D((0,∞))>0D((0,\infty))>0D((0,∞))>0 is added: the paper assumes only nonnegative demands, but with d≡0d\equiv 0d≡0 a state with x0>s0x_0>s_0x0​>s0​ is never left and the claim fails. Policies are deterministic stationary functions of the state; the initial state is any x≥0x\ge 0x≥0, inside or outside X(s)X(s)X(s).

Preamble
import Mathlib
import Definitions.Def_LostSalesOldNew_StateReduction_Model
Formal statement
namespace LostSalesOldNew.StateReduction

open MeasureTheory

theorem claim_1 {L : ℕ} (hL : 0 < L) (s : Fin (L + 1) → ℝ) (hs : IsLevelVector s)
    (z : (Fin L → ℝ) → ℝ) (hz : z ∈ Zs s)
    (D : Measure ℝ) (hD : IsDemand D) (hDpos : 0 < D (Set.Ioi 0))
    (x : Fin L → ℝ) (hx : ∀ i, 0 ≤ x i) :
    haveI := hD.isProb
    ∀ᵐ ds ∂(Measure.infinitePi (fun _ : ℕ => D)),
      ∃ T : ℕ, ∀ t, T ≤ t → traj z x ds t ∈ Xs s := by sorry

end LostSalesOldNew.StateReduction
Source
Zipkin, Old and New Methods for Lost-Sales Inventory Systems, Operations Research 56(5) (2008), Claim 1, p. 1259
Human review
  • Endorsed by Shuze Chen · Oct 6, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 6, 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