Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

§3.1.3, upper-bound claim — the approximation g~λ(q)\tilde g_\lambda(q)g~​λ​(q) of (8) is an upper bound for expected sales gλ(q)g_\lambda(q)gλ​(q)

Proved
FastFashion.Approx.gTilde_upper_bound

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

approximationinventoryp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1poisson-process

Let (Ns)s∈S(N_s)_{s\in\mathcal S}(Ns​)s∈S​ be an independent Poisson family with rates λs>0\lambda_s > 0λs​>0 on a probability space, let T>0T > 0T>0, let S+\mathcal S^+S+ be a nonempty set of major sizes with S−=S∖S+\mathcal S^- = \mathcal S \setminus \mathcal S^+S−=S∖S+, and for every size sss let N(λs)⊆N∪{∞}\mathcal N(\lambda_s) \subseteq \mathbb N \cup \{\infty\}N(λs​)⊆N∪{∞} be a nonempty finite set of tangent indices. Then for every inventory vector q∈NSq \in \mathbb N^{\mathcal S}q∈NS,

gλ(q)=E[G(q)]  ≤  g~λ(q),g_\lambda(q) = \mathbb E[G(q)] \;\le\; \tilde g_\lambda(q),gλ​(q)=E[G(q)]≤g~​λ​(q),

where G(q)G(q)G(q) is the sales of (1) and g~λ\tilde g_\lambdag~​λ​ is the approximation (8) built from the tangents with indices in N(λs)\mathcal N(\lambda_s)N(λs​).

This is the guarantee behind the paper's shipment optimization: the deterministic, piecewise-linear objective that replaces the expected sales in the MIP of §3.2 never underestimates them.

Formalization Note gλg_\lambdagλ​ is defined from GGG by (1), not by formula (2). The tangent coefficients are the index-corrected ones of FastFashion.Approx.Tangents. The paper's rule (7) for choosing N(λs)\mathcal N(\lambda_s)N(λs​) ("bi≈0,0.3T,…b_i \approx 0, 0.3T, \dotsbi​≈0,0.3T,…") is approximate; the statement is made for every nonempty finite choice, which includes the paper's. Nonemptiness of S+\mathcal S^+S+ is an added hypothesis: without a major size the first term of (8), λS+\lambda_{\mathcal S^+}λS+​ times a minimum over ∅\emptyset∅, has no value.

Preamble
import Mathlib
import Definitions.Def_FastFashion_Structure_Model
import Definitions.Def_FastFashion_Approx_Tangents
Formal statement
namespace FastFashion.Approx

open MeasureTheory ProbabilityTheory

/-- §3.1.3, upper-bound claim (Caro–Gallien, p. 13): for every choice of nonempty finite tangent
index sets `𝒩(λ_s) ⊆ ℕ ∪ {∞}` and every inventory vector `q ∈ ℕ^S`, the approximation `g̃_λ(q)`
of (8) is an upper bound for the expected sales `g_λ(q) = 𝔼[G(q)]`. -/
theorem gTilde_upper_bound {S Ω : Type*} [Fintype S] [DecidableEq S] [MeasurableSpace Ω]
    (P : Measure Ω) [IsProbabilityMeasure P] (lam : S → ℝ) (N : S → ℝ → Ω → ℕ)
    (hN : FastFashion.Structure.IsPoissonFamily lam N P) (hlam : ∀ s, 0 < lam s) (T : ℝ) (hT : 0 < T)
    (Sp : Finset S) (hSp : Sp.Nonempty)
    (Nset : S → Finset (WithTop ℕ)) (hNset : ∀ s, (Nset s).Nonempty) (q : S → ℕ) :
    FastFashion.Structure.expectedSales N P Sp T q ≤ gTilde lam T Sp hSp Nset hNset q := by sorry

end FastFashion.Approx
Source
Caro & Gallien, Inventory Management of a Fast-Fashion Retail Network, working paper (August 2, 2007), p. 13, §3.1.3, upper-bound claim (sentence after (8))
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