is a lower bound on the long-run cost of any feasible policy
ProvedInventoryControl.rq_policy_lower_boundThe first half of Proposition 6.1. Under compound Poisson demand with the aperiodicity assumption, let orders be restricted to multiples of a batch quantity and be triggered only by demands, let the lead-time demand distribution be that of the demand over an interval of length , let be its holding-plus-shortage cost rate with , and let minimize . Then for every ordering rule whatsoever and every initial position , almost surely,
where is the inventory position under .
The argument is pathwise: every policy's position is congruent modulo to , so by Lemma 6.1 its cost rate is at least at every instant, and the long-run average of is . No measurability, adaptedness or stability of the rule is needed; a rule under which the position drifts to infinity simply has infinite average cost.
Formalization Note "" is written without a liminf: for every
the average eventually exceeds . The lead-time demand law
is tied to the process by the hypothesis that is the distribution of the demand in
.
import Definitions.Def_InventoryControl_rqPolicy
namespace InventoryControl
theorem rq_policy_lower_bound {Ω : Type*} [MeasurableSpace Ω] {P : MeasureTheory.Measure Ω}
[MeasureTheory.IsProbabilityMeasure P] (X : CompoundPoissonDemand P) (D : DiscreteDemand)
(h b1 L : ℝ) (hh : 0 < h) (hb : 0 < b1) (hL : 0 ≤ L)
(hD : ∀ j : ℕ, D.p j = P.real {ω | X.cumDemand (X.count L ω) ω = j})
(Q : ℕ) (hQ : 0 < Q) (R : ℤ)
(hR : ∀ y : ℤ, windowCost D h b1 Q R ≤ windowCost D h b1 Q y) (y0 : ℤ)
(m : ℕ → Ω → ℕ) :
∀ᵐ ω ∂P, ∀ ε : ℝ, 0 < ε →
∀ᶠ T in Filter.atTop, windowCost D h b1 Q R / Q - ε ≤ avgCost X D h b1 L Q y0 m T ω := by sorry
end InventoryControl
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.