Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Lemma 2 — |v(x + h eⁱ, Y) − v(x, Y)| ≤ h r̄ with r̄ = max_j r^j

Proved
RLPBidPrice.Directional.piValue_lipschitz

by mikedeng1 · Oct 3, 2026 · Mathlib 0df444a (Lean v4.33.1)

linear-programmingp2o-batch-rm-cafterp2o-gran-per-chapterp2o-plan-paperp2o-v1revenue-management

Let AAA be a 0/1 leg–itinerary incidence matrix with nonzero columns, r≥0r \ge 0r≥0 the fares, x≥0x \ge 0x≥0 a capacity vector and Y≥0Y \ge 0Y≥0 a demand vector, and let v(x,Y)=max⁡{r⊤y:Ay≤x, 0≤y≤Y}v(x, Y) = \max\{r^\top y : Ay \le x,\ 0 \le y \le Y\}v(x,Y)=max{r⊤y:Ay≤x, 0≤y≤Y}. For every leg iii and every h≥0h \ge 0h≥0,

∣v(x+hei,Y)−v(x,Y)∣≤h rˉ,rˉ=max⁡jrj,|v(x + h e^i, Y) - v(x, Y)| \le h\, \bar r, \qquad \bar r = \max_{j} r^j,∣v(x+hei,Y)−v(x,Y)∣≤hrˉ,rˉ=jmax​rj,

where eie^iei is the iii-th unit vector.

This is the paper's Lipschitz bound (Lemma 2: "vk(x,Y)v_k(x, Y)vk​(x,Y) satisfies the Lipschitz condition in Lemma 1"): a constant rˉ\bar rrˉ in each coordinate direction, uniform in YYY. In the nondifferentiable case it is what makes the difference quotients of vvv uniformly bounded.

Formalization Note. rˉ\bar rrˉ is written sup⁡jrj\sup_j r^jsupj​rj (⨆ j, r j), which is max⁡jrj\max_j r^jmaxj​rj for n≥1n \ge 1n≥1 and 000 when there are no itineraries. The paper's "for all xxx" is restricted to x≥0x \ge 0x≥0: for capacities with a negative entry the LP is infeasible and has no value. Fares r≥0r \ge 0r≥0 are an implicit hypothesis (revenues): if every rj<0r^j < 0rj<0 then v≡0v \equiv 0v≡0 and the bound hrˉ<0h \bar r < 0hrˉ<0 fails. The time index kkk is dropped.

Preamble
import Mathlib
import Definitions.Def_RLPBidPrice_Directional_Model

open MeasureTheory Matrix
Formal statement
namespace RLPBidPrice.Directional

/-- Lemma 2 (p. 210) and the bound preceding it: raising the capacity of one leg by `h ≥ 0`
changes the perfect-information LP value by at most `h · r̄`, `r̄ = max_j r^j`. -/
theorem piValue_lipschitz {m n : ℕ} (A : Matrix (Fin m) (Fin n) ℝ) (r : Fin n → ℝ)
    (hA : IsIncidence A) (hr : 0 ≤ r) (x : Fin m → ℝ) (hx : 0 ≤ x)
    (y₀ : Fin n → ℝ) (hy₀ : 0 ≤ y₀) (i : Fin m) (h : ℝ) (hh : 0 ≤ h) :
    |piValue A r (x + h • Pi.single i 1) y₀ - piValue A r x y₀| ≤ h * ⨆ j, r j := by sorry

end RLPBidPrice.Directional
Source
Talluri, van Ryzin, A Randomized Linear Programming Method for Computing Network Bid Prices, Transp. Sci. 33(2), 1999, p. 210, Lemma 2 and the sentence preceding it
Read-back

What the Lean code literally says, in plain math · claude-opus-5-5

Setting. Let m,n∈Nm, n \in \mathbb{N}m,n∈N, let AAA be a real m×nm \times nm×n matrix, and let r∈Rnr \in \mathbb{R}^nr∈Rn. All vector inequalities are componentwise.

Hypotheses.

  • AAA is an incidence matrix: every entry is 000 or 111, and every column has at least one nonzero entry.
  • r≥0r \ge 0r≥0.
  • x∈Rmx \in \mathbb{R}^mx∈Rm with x≥0x \ge 0x≥0.
  • y0∈Rny_0 \in \mathbb{R}^ny0​∈Rn with y0≥0y_0 \ge 0y0​≥0.
  • i∈{1,…,m}i \in \{1,\dots,m\}i∈{1,…,m} is an index.
  • h∈Rh \in \mathbb{R}h∈R with h≥0h \ge 0h≥0.

Let eie_iei​ be the iii-th standard unit vector of Rm\mathbb{R}^mRm. Define

v(x,y0)=sup⁡{ r⋅y:Ay≤x, 0≤y≤y0 },v(x,y_0) = \sup\{\, r\cdot y : Ay \le x,\ 0 \le y \le y_0 \,\},v(x,y0​)=sup{r⋅y:Ay≤x, 0≤y≤y0​},

with the real-number convention that this is 000 when the set is empty or unbounded above.

Conclusion.

∣ v(x+h ei, y0)−v(x,y0) ∣  ≤  h⋅sup⁡jrj,\big|\, v(x + h\, e_i,\ y_0) - v(x, y_0) \,\big| \;\le\; h \cdot \sup_{j} r_j ,​v(x+hei​, y0​)−v(x,y0​)​≤h⋅jsup​rj​,

where sup⁡jrj\sup_j r_jsupj​rj​ is the supremum over j∈{1,…,n}j \in \{1,\dots,n\}j∈{1,…,n} in R\mathbb{R}R. For n≥1n \ge 1n≥1 this is max⁡jrj\max_j r_jmaxj​rj​.

Degenerate cases.

  • Under these hypotheses both LPs are feasible (y=0y = 0y=0 works), so neither value is the junk value.
  • If h=0h = 0h=0, the claim is 0≤00 \le 00≤0.
  • If n=0n = 0n=0: the empty supremum is 000 and both values are 000, so the claim is 0≤00 \le 00≤0.
  • If m=0m = 0m=0: no index iii exists, so the statement is vacuous. The incidence hypothesis would also then force n=0n = 0n=0.
Human review
  • Endorsed by Shuze Chen · Oct 21, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 21, 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