Brown's concavity theorem: with log-convex survival, is concave on
ProvedOWMR.renewal_concave_dfrapproximation-algorithmsinventoryinventory-theoryoperations-researchrenewal-theorystochastic-control
Let be a probability law on with positive mean, and let be its survival function. Suppose is log-convex on : for and ,
Let be the renewal measure, including the initial renewal at zero (renewalMeasure). Then the renewal function is concave on .
The law may have an atom at zero. Dividing by shows that the hypothesis is exactly log-convexity of the survival function conditional on positive demand, the DFR property of Appendix D.
Preamble
import Definitions.Def_owmr_renewal open MeasureTheory open scoped ENNReal NNReal
Formal statement
namespace OWMR
theorem renewal_concave_dfr (ρ : Measure ℝ≥0) [IsProbabilityMeasure ρ]
(hρ : 0 < ∫⁻ x, (x : ℝ≥0∞) ∂ρ)
(hlc : ∀ x y η : ℝ, 0 ≤ x → 0 ≤ y → 0 ≤ η → η ≤ 1 →
(ρ {z : ℝ≥0 | η * x + (1 - η) * y < (z : ℝ)}).toReal
≤ (ρ {z : ℝ≥0 | x < (z : ℝ)}).toReal ^ η * (ρ {z : ℝ≥0 | y < (z : ℝ)}).toReal ^ (1 - η)) :
ConcaveOn ℝ (Set.Ici 0) (fun x : ℝ => (renewalMeasure ρ (Set.Icc 0 x)).toReal) := 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), Appendix D, paragraph before the proof of Theorem D.1, p. E.25: for absolutely continuous DFR , is concave on (Brown 1980; Yu 2011), and with .