THEOREM 5 (20) — D = Σ p_j D_j for k equal subintervals of a linear rate
ProvedNHPPArrivals.LinearRate.mixture_degreedegree-of-nonhomogeneitylinear-arrival-ratemixturep2o-batch-p200ap2o-gran-per-chapterp2o-plan-paperp2o-v1
Let on with , , and or . Divide into subintervals of length , with subinterval conditional cdfs and weights as in LEMMA 1, and let be the conditional cdf of the combined data. Then every supremum below is attained on , and
The degree of nonhomogeneity of the combined data is thus the weighted average of the degrees of the subintervals, with the arrival shares as weights. This is the first display of THEOREM 5, for one fixed , without the closed forms (21)–(22).
Formalization Note is §3.3's standing assumption, and " or " excludes the identically zero rate (§3.2). covers the paper's two cases and .
Preamble
import Mathlib import Definitions.Def_NHPPArrivals_LinearRate_ConditionalCdf
Formal statement
namespace NHPPArrivals.LinearRate
theorem mixture_degree (a b T : ℝ) (k : ℕ) (hT : 0 < T) (ha : 0 ≤ a) (hb : 0 ≤ b)
(hab : 0 < a ∨ 0 < b) (hk : 1 ≤ k) :
IsGreatest ((fun t => |mixCdf (linRate a b) T k t - t|) '' Set.Icc (0:ℝ) 1)
(degree (mixCdf (linRate a b) T k)) ∧
(∀ j ∈ Finset.Icc 1 k,
IsGreatest ((fun t => |subCdf (linRate a b) T k j t - t|) '' Set.Icc (0:ℝ) 1)
(degree (subCdf (linRate a b) T k j))) ∧
degree (mixCdf (linRate a b) T k) =
∑ j ∈ Finset.Icc 1 k,
weight (linRate a b) T k j * degree (subCdf (linRate a b) T k j) := by sorry
end NHPPArrivals.LinearRate
Source
Kim and Whitt, Are call center and hospital arrivals well modeled by nonhomogeneous Poisson processes?, Manufacturing Service Oper. Management 16(3), 2014, p. 473, THEOREM 5, (20)
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.