Eq. (26): reduction of the general-setup optimum to the zero-setup optimum
ProvedLostSalesRQ.operating_transformapproximation-guaranteeinventorylost-salesoperations-research
Fix , and .
- For every with ,
- (the case ).
A fraction of holding cost and all setup costs pay for the service credit of Lemma J.1; the remaining costs form a zero-setup problem with an adjusted penalty.
Preamble
import Definitions.Def_LostSalesRQ_model
Formal statement
namespace LostSalesRQ
theorem operating_transform (μ a k : ℝ) (hμ : 0 ≤ μ) (ha : 0 ≤ a) (hk : 0 ≤ k) :
(∀ θ : ℝ, 0 ≤ θ → θ < 1 → Real.sqrt (2 * k * θ) ≤ a →
ENNReal.ofReal (Real.sqrt (2 * k * θ)) +
ENNReal.ofReal (1 - θ) * OPTk μ ((a - Real.sqrt (2 * k * θ)) / (1 - θ)) 0 ≤
OPTk μ a k) ∧
ENNReal.ofReal (min a (Real.sqrt (2 * k))) ≤ OPTk μ a k := by sorry
end LostSalesRQSource
M. Hu, "A 1.932 Guarantee for (r,Q) Policies in Continuous-Review Lost-Sales Inventory Systems", arXiv:2610.34576, 2026; Section 4.6.2, eq. (26) and the following paragraph (p. 22); Appendix J.1, eq. (J.2) (p. A22).
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.