Sec. 2.3, p. 7 — ϑ > 0 and γ ∈ [0, 1)
ProvedLostSalesExpGap.ConstOrder.vartheta_pos_gamma_lt_onelarge-deviationslost-salesp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1
In the lost-sales model with strictly positive demand variance, let be the best constant order, for , , and the supremum of the minimizers of on ( if the infimum is not attained). Then
These are what make an exponentially decaying factor and , finite in Theorem 1.
Formalization Note is stated for the case where the set of minimizers is nonempty and bounded (then is its supremum); in the other cases by definition.
Preamble
import Mathlib import Definitions.Def_LostSalesExpGap_ConstOrder_ConstantOrder open MeasureTheory ProbabilityTheory open scoped ENNReal NNReal BigOperators
Formal statement
namespace LostSalesExpGap.ConstOrder
/-- Sec. 2.3, p. 7: `ϑ > 0` and `γ ∈ [0, 1)`. When the minimizers of `φ` on `[0, ∞)` form a
nonempty bounded set, their supremum `ϑ` is positive; otherwise `ϑ = ∞`. -/
theorem vartheta_pos_gamma_lt_one (P : CappedBaseStock.DemandLaw)
(c : CappedBaseStock.Parameters)
(hP : LostSalesConstantOrder.LargeLeadTime.NonDeterministic P) :
0 ≤ gamma P c ∧ gamma P c < 1 ∧
((phiMinimizers P c).Nonempty → BddAbove (phiMinimizers P c) →
0 < sSup (phiMinimizers P c)) := by sorry
end LostSalesExpGap.ConstOrder
Source
Xin, Goldberg, Optimality gap of constant-order policies decays exponentially in the lead time for lost sales models, arXiv:1409.1499v2, Sec. 2.3, p. 7
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.