Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

pp. 23–24 — the greedy fill ẑ_i(ϵ_i) is optimal for the continuous knapsack (11) and has at most one fractional component

Proved
NestedLogitVariants.PartialCompetitive.zhat_optimal

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

greedyknapsacklinear-programmingp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1

Fix a nest iii of a nested logit instance (revenues ordered ri1≥⋯≥rinr_{i1} \ge \dots \ge r_{in}ri1​≥⋯≥rin​, weights vij>0v_{ij} > 0vij​>0) and a capacity ϵi≥0\epsilon_i \ge 0ϵi​≥0. Let z^i(ϵi)\hat z_i(\epsilon_i)z^i​(ϵi​) be the greedy solution of the continuous knapsack problem (11): fill the capacity with the eligible products (vij≤ϵiv_{ij} \le \epsilon_ivij​≤ϵi​) in revenue order, each fully while it fits and the next one fractionally. Then:

  1. z^i(ϵi)\hat z_i(\epsilon_i)z^i​(ϵi​) is feasible for (11): ∑jvijz^ij(ϵi)≤ϵi\sum_j v_{ij}\hat z_{ij}(\epsilon_i) \le \epsilon_i∑j​vij​z^ij​(ϵi​)≤ϵi​ and 0≤z^ij(ϵi)≤1(vij≤ϵi)0 \le \hat z_{ij}(\epsilon_i) \le \mathbf 1(v_{ij} \le \epsilon_i)0≤z^ij​(ϵi​)≤1(vij​≤ϵi​);
  2. it is optimal for (11): for every feasible ziz_izi​,
∑j∈Nrijvijzij≤∑j∈Nrijvijz^ij(ϵi)=K^i(ϵi);\sum_{j \in N} r_{ij} v_{ij} z_{ij} \le \sum_{j \in N} r_{ij} v_{ij} \hat z_{ij}(\epsilon_i) = \hat K_i(\epsilon_i);j∈N∑​rij​vij​zij​≤j∈N∑​rij​vij​z^ij​(ϵi​)=K^i​(ϵi​);
  1. at most one component z^ij(ϵi)\hat z_{ij}(\epsilon_i)z^ij​(ϵi​) lies strictly between 000 and 111.

Optimality of the greedy fill identifies K^i(ϵi)\hat K_i(\epsilon_i)K^i​(ϵi​), and the single fractional component is what makes the case analysis of the proof of Theorem 10 possible.

Preamble
import Mathlib
import Definitions.Def_NestedLogitVariants_PartialCompetitive_Model
import Definitions.Def_NestedLogitVariants_PartialCompetitive_Knapsack
Formal statement
namespace NestedLogitVariants.PartialCompetitive

/-- pp. 23–24: the greedy fill `ẑ_i(ε)` is feasible for (11), optimal for (11), and has at most
one fractional component. -/
theorem zhat_optimal {ι : Type*} [Fintype ι] [DecidableEq ι] {n : ℕ} (I : Instance ι n)
    (hI : I.Standing)
    (i : ι) (ε : ℝ) (hε : 0 ≤ ε) :
    feas11 I i ε (zhat I i ε) ∧
      (∀ z : Fin n → ℝ, feas11 I i ε z →
        ∑ j, I.r i j * I.v i j * z j ≤ ∑ j, I.r i j * I.v i j * zhat I i ε j) ∧
      (∀ j k : Fin n, zhat I i ε j ∈ Set.Ioo (0 : ℝ) 1 → zhat I i ε k ∈ Set.Ioo (0 : ℝ) 1 →
        j = k) := by sorry

end NestedLogitVariants.PartialCompetitive
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. 23–24, the paragraph after display (11)
Human review
  • Endorsed by Shuze Chen · Oct 5, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 5, 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