Theorem 4, p. 939, with §3, p. 938 — the optimal lost-sales cost f̄_t(v) is convex on V
ProvedZipkinLostSales.Variability.fbar_convexOnConsider the lost-sales model of Zipkin (2008) in the transformed state, with lead time , costs , discount factor , and one-period demand law , a probability measure on with finite mean. Let be the state space and the optimal cost functions of recursion (1)–(2).
Then for all , is convex on :
The paper obtains this from Theorem 4 ( is L-convex) and its remark in §3 that L-convexity implies ordinary convexity. In the proof of Theorem 11, convexity of the continuation is what makes jointly convex in .
Formalization Note The optimal cost is indexed by the number of periods to go , with at ; "for all " is "for all ". The statement is for the model's only, not a generic "L-convex implies convex", which fails without regularity. The conditions , read the paper's "unit cost" and "discount rate"; finite mean of demand is added so that is finite.
import Mathlib import Definitions.Def_ZipkinLostSales_Variability_Model open MeasureTheory
namespace ZipkinLostSales.Variability
/-- Theorem 4 (Zipkin 2008, p. 939) with the remark of §3 (p. 938) that L♮-convexity implies
ordinary convexity, for the model's value functions: for one demand law `μ` and every number of
periods to go `k`, `f̄` is convex on the state space `V`. -/
theorem fbar_convexOn {L : ℕ} (hL : 0 < L) (c hh p γ : ℝ) (hc : 0 ≤ c) (hhh : 0 ≤ hh)
(hp : 0 ≤ p) (hγ0 : 0 < γ) (hγ1 : γ ≤ 1) (μ : Measure ℝ) [IsProbabilityMeasure μ]
(hμ : μ (Set.Iio 0) = 0) (hμint : Integrable id μ) :
∀ k : ℕ, ConvexOn ℝ (ZipkinLostSales.LNatural.V L) (ZipkinLostSales.LNatural.fbar c hh p γ μ k) := by sorry
end ZipkinLostSales.Variability
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.