Proposition 5.1: under the policy the inventory position is uniform on
ProvedInventoryControl.rq_position_uniformProposition 5.1 in its ergodic form. Under compound Poisson demand whose demand sizes are not all multiples of some integer larger than one, an policy with started from any initial inventory position spends, in the long run, a fraction of the time at each of the positions : for every such , almost surely
where is the position after the -th demand under the policy.
The book proves the proposition for the embedded chain at demand epochs by showing that the transition matrix is doubly stochastic (Eq. 5.33-5.34), which makes the uniform distribution stationary, and irreducibility, which comes from the aperiodicity assumption, makes it the unique one. Here the statement is about time averages, which is how the proposition is used: the cost of the policy is the average of over the band. The initial position may lie above the band; the policy enters the band after finitely many demands and the initial segment does not affect the limit.
import Definitions.Def_InventoryControl_rqPolicy
namespace InventoryControl
theorem rq_position_uniform {Ω : Type*} [MeasurableSpace Ω] {P : MeasureTheory.Measure Ω}
[MeasureTheory.IsProbabilityMeasure P] (X : CompoundPoissonDemand P) (R : ℤ) (Q : ℕ)
(hQ : 0 < Q) (y0 : ℤ) (j : ℤ) (hj1 : R + 1 ≤ j) (hj2 : j ≤ R + Q) :
∀ᵐ ω ∂P, Filter.Tendsto
(fun T : ℝ => (1 / T) * ∫ t in (0 : ℝ)..T,
(if rqIP X R Q y0 (X.count t ω) ω = j then (1 : ℝ) else 0))
Filter.atTop (nhds (1 / (Q : ℝ))) := by sorry
end InventoryControl
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.