Existence of the myopic basestock levels (smallest minimizers of )
ProvedSongZipkinFluct.Monotone.yplus_existsbasestock-policyinventorymarkov-modulated-demand
Assume the standing hypotheses of the model and Assumption 1 (). For every world state , the myopic cost
has a smallest global minimizer over the integers. Here is the discounted expected inventory cost at the end of the lead time.
The paper (p. 356) defines as the smallest value of minimizing ; its existence follows from convexity of (Lemma 1) and coercivity under Assumption 1 (Lemma 2), p. 355. This is the existence part used in Theorem 8.
Preamble
import Mathlib import Definitions.Def_SongZipkinFluct_Monotone_Model
Formal statement
namespace SongZipkinFluct.Monotone
theorem yplus_exists {I : Type} [Countable I] [Nonempty I] [DecidableEq I]
(M : Model I) (hM : M.Standing) (hA : M.Assumption1) :
∃ yplus : I → ℤ, ∀ i, IsSmallestMinimizer (M.Gplus i) (yplus i) := by sorry
end SongZipkinFluct.Monotone
Source
Song and Zipkin, Inventory Control in a Fluctuating Demand Environment, Oper. Res. 41(2):351–370 (1993), DOI 10.1287/opre.41.2.351, p. 355 (Lemmas 1–2) and p. 356 (definition of the myopic level y⁺(i))