Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

(20), Appendix A.1, Case 1.a, pp. 41–42 — the inequality with α¹_ik = R_i(N_i,k−1)/R_i(N_ik)

Proved
NestedLogitVariants.Synergistic.ineq_20

by mikedeng1 · Oct 5, 2026 · Mathlib 0df444a (Lean v4.33.1)

assortment-optimizationnested-logitp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1revenue-management

Let (x^,y^)(\hat x, \hat y)(x^,y^​) be an optimal solution of (4) over the nested-by-revenue assortments {Nij:j∈N+}\{N_{ij} : j \in N_+\}{Nij​:j∈N+​} for an instance with fully-captured nests and positive revenues. Fix a nest iii, an index kkk with 2≤k≤n2 \le k \le n2≤k≤n and ρ∈[0,1]\rho \in [0, 1]ρ∈[0,1], write Rik′=∑j=1k′rijvijR_{ik'} = \sum_{j=1}^{k'} r_{ij} v_{ij}Rik′​=∑j=1k′​rij​vij​, qik′=∑j=1k′vijq_{ik'} = \sum_{j=1}^{k'} v_{ij}qik′​=∑j=1k′​vij​ and

αik1=Ri(Ni,k−1)Ri(Nik).\alpha^1_{ik} = \frac{R_i(N_{i,k-1})}{R_i(N_{ik})}.αik1​=Ri​(Nik​)Ri​(Ni,k−1​)​.

Then

αik1 y^i≥(qi,k−1+vikρ)γi[Ri,k−1+rikvikρqi,k−1+vikρ−αik1 x^].(20)\alpha^1_{ik}\, \hat y_i \ge (q_{i,k-1} + v_{ik}\rho)^{\gamma_i} \left[\frac{R_{i,k-1} + r_{ik} v_{ik} \rho}{q_{i,k-1} + v_{ik}\rho} - \alpha^1_{ik}\, \hat x\right]. \tag{20}αik1​y^​i​≥(qi,k−1​+vik​ρ)γi​[qi,k−1​+vik​ρRi,k−1​+rik​vik​ρ​−αik1​x^].(20)

The right side is the objective of (8) at the fractional nested-by-revenue vector with zi1=⋯=zi,k−1=1z_{i1} = \dots = z_{i,k-1} = 1zi1​=⋯=zi,k−1​=1, zik=ρz_{ik} = \rhozik​=ρ, evaluated at αik1x^\alpha^1_{ik} \hat xαik1​x^; (20) is the first of the two bounds combined in the proof of Theorem 7.

Formalization Note The standing assumptions are those of the model (see the Model definition) and of §4: every nest is fully captured, vi0=0v_{i0} = 0vi0​=0, and γi>1\gamma_i > 1γi​>1 for some nest iii (p. 16). The positive revenues rij>0r_{ij} > 0rij​>0 are the disclosed pin of this mission: they keep every denominator Ri(Nij)R_i(N_{ij})Ri​(Nij​), j≥1j \ge 1j≥1, of (6) positive. Product kkk of the page (indexed from 111) is ⟨k - 1, _⟩ : Fin n; Ri,k−1R_{i,k-1}Ri,k−1​ and qi,k−1q_{i,k-1}qi,k−1​ are the sums over Ni,k−1N_{i,k-1}Ni,k−1​.

Preamble
import Mathlib
import Definitions.Def_NestedLogitVariants_Synergistic_Relaxation
Formal statement
namespace NestedLogitVariants.Synergistic

/-- Inequality (20), Appendix A.1, Case 1.a, pp. 41–42: for `2 ≤ k ≤ n` and `ρ ∈ [0, 1]`, with
`α¹_{ik} = R_i(N_{i,k−1}) / R_i(N_{ik})`,
`α¹_{ik} ŷ_i ≥ (q_{i,k−1} + v_{ik} ρ)^{γ_i} [(R_{i,k−1} + r_{ik} v_{ik} ρ)/(q_{i,k−1} + v_{ik} ρ) − α¹_{ik} x̂]`.
Product `k` (1-based) is `⟨k - 1, _⟩ : Fin n`. -/
theorem ineq_20 {ι : Type*} [Fintype ι] [DecidableEq ι] {n : ℕ} (I : Instance ι n)
    (hI : I.Standing) (hfc : ∀ i, I.vnp i = 0) (hsyn : ∃ i, 1 < I.γ i)
    (hr : ∀ i j, 0 < I.r i j)
    (xh : ℝ) (yh : ι → ℝ) (hopt : LP4Optimal I (fun _ => {S | ∃ j ≤ n, S = nbr n j}) xh yh)
    (i : ι) (k : ℕ) (hk2 : 2 ≤ k) (hkn : k ≤ n) (ρ : ℝ) (hρ : ρ ∈ Set.Icc (0 : ℝ) 1) :
    (R I i (nbr n (k - 1)) / R I i (nbr n k)) * yh i ≥
      (qsum I i (k - 1) + I.v i ⟨k - 1, by omega⟩ * ρ) ^ I.γ i *
        ((Rsum I i (k - 1) + I.r i ⟨k - 1, by omega⟩ * I.v i ⟨k - 1, by omega⟩ * ρ) / (qsum I i (k - 1) + I.v i ⟨k - 1, by omega⟩ * ρ) -
          (R I i (nbr n (k - 1)) / R I i (nbr n k)) * xh) := by sorry

end NestedLogitVariants.Synergistic
Source
Davis, Gallego, Topaloglu, Assortment optimization under variants of the nested logit model, revised manuscript of June 18, 2013 (published Oper. Res. 62(2), 2014, DOI 10.1287/opre.2014.1256), pp. 41–42, Appendix A.1, Case 1.a, display (20)
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