Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem D.1 (DFR Demand): C(πDFR(θ))≤γλ OPTC(\pi_{\rm DFR}(\theta))\le\gamma_\lambda\,\mathrm{OPT}C(πDFR​(θ))≤γλ​OPT, γλ=1+λ/2+1+λ2/4\gamma_\lambda=1+\lambda/2+\sqrt{1+\lambda^2/4}γλ​=1+λ/2+1+λ2/4​, λ=2−2\lambda=2-\sqrt2λ=2−2​

Proved
OWMR.dfr_guarantee

by Ming Hu · Nov 17, 2026 · Mathlib 0df444a (Lean v4.33.1)

approximation-algorithmsinventoryinventory-theoryoperations-researchrenewal-theorystochastic-control

Let λ=2−2\lambda=2-\sqrt2λ=2−2​ and γλ=1+λ/2+1+λ2/4≈2.335\gamma_\lambda=1+\lambda/2+\sqrt{1+\lambda^2/4}\approx2.335γλ​=1+λ/2+1+λ2/4​≈2.335. Consider an OWMR network with n≥1n\ge1n≥1 retailers under the standing assumptions of Section 3: demand vectors are i.i.d. over periods and may be dependent across retailers, and the lead times L0,Li≥0L_0,L_i\ge0L0​,Li​≥0 are arbitrary finite integers. Suppose Hi≥hH_i\ge hHi​≥h, p≥hp\ge hp≥h and Ki≤(γλ−1)K0K_i\le(\gamma_\lambda-1)K_0Ki​≤(γλ​−1)K0​ for every retailer, and that each retailer with Ki>0K_i>0Ki​>0 has an absolutely continuous DFR distribution conditional on positive demand (DFRPositive). Let θ=1−1/γλ\theta=1-1/\gamma_\lambdaθ=1−1/γλ​, and let ccc be any choices satisfying Definition D.1 (IsDFRChoice N θ c). Then, from every valid initial state xxx, the policy πDFR(θ)\pi_{\rm DFR}(\theta)πDFR​(θ), the FIFO policy fifoPolicy N x c, is nonanticipative and feasible, and

C(πDFR(θ))≤γλ OPT,C\bigl(\pi_{\rm DFR}(\theta)\bigr)\le\gamma_\lambda\,\mathrm{OPT},C(πDFR​(θ))≤γλ​OPT,

where CCC is the long-run average cost (2) and OPT\mathrm{OPT}OPT is the infimum of CCC over all nonanticipative, feasible, measurable policies, which may randomize independently of demand.

Preamble
import Definitions.Def_owmr_dfr_policy

open scoped ENNReal
Formal statement
namespace OWMR
theorem dfr_guarantee {n : ℕ} (hn : 0 < n) (N : Network n) (hN : N.Standing)
    (hH : ∀ i, N.h ≤ N.H i) (hp : N.h ≤ N.p) (hK : ∀ i, N.K i ≤ (dfrGamma - 1) * N.K0)
    (hD : ∀ i, 0 < N.K i → N.DFRPositive i)
    (x : InitState n) (hx : x.Valid) (c : FIFOChoice n) (hc : IsDFRChoice N (1 - 1 / dfrGamma) c) :
    (fifoPolicy N x c).toPolicy.Nonanticipative ∧ (fifoPolicy N x c).toPolicy.Feasible N x ∧
      (fifoPolicy N x c).avgCost N x ≤ ENNReal.ofReal dfrGamma * OPT N x := by sorry
end OWMR
Source
M. Hu, "A 5/2 Guarantee for Echelon Stock (s,S) Policies in One-Warehouse Multi-Retailer Systems", working paper (manuscript of November 2026), Appendix D, Theorem D.1 (DFR Demand), p. E.24, with Definition D.1 (DFR Policy), p. E.23, and the proof of Theorem D.1, Steps 1-5 and (D.5)-(D.9), pp. E.25-E.27.
Human review
  • Endorsed by Shuze Chen · Nov 17, 2026

    Confirmed by the moderator at approval.

  • Endorsed by Ming Hu · Nov 17, 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