Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

The reduced position yt′y_t'yt′​ spends a fraction 1/Q1/Q1/Q of the time at each point of the band

Proved
InventoryControl.reduced_position_uniform

by naimengye · Sep 24, 2026 · Mathlib 0df444a (Lean v4.33.1)

ergodic-theoryinventory-controlmarkov-chainoperations-researchstochastic-processes

Under compound Poisson demand whose demand sizes are not all multiples of some integer larger than one, fix a batch quantity Q≥1Q \ge 1Q≥1, a reorder point RRR and an initial position y0y_0y0​, and let yt′∈{R+1,…,R+Q}y_t' \in \{R+1, \dots, R+Q\}yt′​∈{R+1,…,R+Q} be the reduction modulo QQQ of y0−SN(t)y_0 - S_{N(t)}y0​−SN(t)​, where SN(t)S_{N(t)}SN(t)​ is the total demand by time ttt. Then for every j∈{R+1,…,R+Q}j \in \{R+1, \dots, R+Q\}j∈{R+1,…,R+Q}, almost surely,

1T∫0T1{yt′=j} dt  ⟶  1Q(T→∞).\frac{1}{T}\int_0^T \mathbf{1}\{y_t' = j\}\,\mathrm{d}t \;\longrightarrow\; \frac{1}{Q} \qquad (T \to \infty).T1​∫0T​1{yt′​=j}dt⟶Q1​(T→∞).

This is the ergodic form of "the steady state distribution is uniform". The process yt′y_t'yt′​ is the same for every ordering rule, because every policy's position is congruent to y0−SN(t)y_0 - S_{N(t)}y0​−SN(t)​ modulo QQQ, and it is a continuous-time random walk on the QQQ residues driven by the demand sizes; the aperiodicity assumption makes that walk irreducible, and the uniform distribution is stationary because the walk is doubly stochastic (the argument of Proposition 5.1, Eq. 5.33-5.34). The long-run time average of any function of yt′y_t'yt′​ is then its average over the band.

Preamble
import Definitions.Def_InventoryControl_rqPolicy
Formal statement
namespace InventoryControl

theorem reduced_position_uniform {Ω : Type*} [MeasurableSpace Ω] {P : MeasureTheory.Measure Ω}
    [MeasureTheory.IsProbabilityMeasure P] (X : CompoundPoissonDemand P) (R : ℤ) (Q : ℕ)
    (hQ : 0 < Q) (y0 : ℤ) (j : ℤ) (hj1 : R + 1 ≤ j) (hj2 : j ≤ R + Q) :
    ∀ᵐ ω ∂P, Filter.Tendsto
      (fun T : ℝ => (1 / T) * ∫ t in (0 : ℝ)..T,
          (if reduceToBand R Q (y0 - (X.cumDemand (X.count t ω) ω : ℤ)) = j then (1 : ℝ) else 0))
      Filter.atTop (nhds (1 / (Q : ℝ))) := by sorry

end InventoryControl
Source
Sven Axsäter, Inventory Control, 3rd ed., Springer 2015, DOI 10.1007/978-3-319-15729-0, p. 115, Sect. 6.2.1: 'The demand sizes are independent, so the different y't can be seen as a Markov chain with the finite state space {R + 1, R + 2, ..., R + Q}. The steady state distribution can be shown to be uniform'; the argument is that of Proposition 5.1, p. 74
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).

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