Scaling a setup cost by scales the value by at most
ProvedOWMR.slValue_setup_scaleapproximation-algorithmsinventoryinventory-theoryoperations-researchstochastic-control
Let be a probability law of demand on and a nonnegative one-period holding and backlog cost, and write for the optimal long-run average cost of the single-location problem with i.i.d. demands of law and setup cost . For every real and every ,
The source uses this in Step 2 of the proof of Theorem 1, with , when .
Formalization Note is slValue ρ f K, and the factor enters through ENNReal.ofReal.
Preamble
import Definitions.Def_owmr_single_location open MeasureTheory open scoped ENNReal NNReal
Formal statement
namespace OWMR
theorem slValue_setup_scale (ρ : Measure ℝ≥0) [IsProbabilityMeasure ρ] (f : ℝ → ℝ)
(hf : ∀ y, 0 ≤ f y) (K : ℝ) {a : ℝ} (ha : 1 ≤ a) :
slValue ρ f (a * K) ≤ ENNReal.ofReal a * slValue ρ f K := by sorry
end OWMRSource
M. Hu, "A 5/2 Guarantee for Echelon Stock (s,S) Policies in One-Warehouse Multi-Retailer Systems", working paper (manuscript of November 2026), Section 5.4, Step 2 of the proof of Theorem 1 ("multiplying a setup cost by a >= 1 multiplies its value by at most a"), p. 26.