§2.2.3, Corollary (3)(b), p. 539 — with λ = Σ_k λ_k, L = Σ_k L_k and W = Σ_k (λ_k/λ) W_k, L = λW
ProvedLittleLaw50.FiniteWindow.corollary_3blittles-lawp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1queueing-theory
In the setting of Corollary (3)(a): one sample path observed over with , items with , divided into mutually exclusive classes , and , , the LL parameters of class computed from the class- items alone. Let
Then
The aggregate parameters built from the class-wise ones satisfy Little's Law again, with the arrival-weighted average of the class waits.
Formalization Note , and are computed from the sample path (they are not free numbers). If (no class has an item in the system over ), Lean's is and both sides are .
Preamble
import Mathlib import Definitions.Def_KellyStochasticNetworks_Migration import Definitions.Def_LittleLaw50_FiniteWindow_Window
Formal statement
namespace LittleLaw50.FiniteWindow
/-- **§2.2.3, Corollary (3)(b)** (Little 2011, p. 539). With `L_k, λ_k, W_k` the LL parameters of
class `k`, put `λ = Σ_k λ_k`, `L = Σ_k L_k` and `W = Σ_k (λ_k/λ) W_k`. Then `L = λ W`. -/
theorem corollary_3b {M K : ℕ} (a d : Fin M → ℝ) (c : Fin M → Fin K) (T : ℝ) (hT : 0 < T)
(had : ∀ i, a i ≤ d i) :
(∑ k, Lw (classItems c k) a d T)
= (∑ k, lamw (classItems c k) a d T)
* ∑ k, (lamw (classItems c k) a d T / ∑ k', lamw (classItems c k') a d T)
* Ww (classItems c k) a d T := by sorry
end LittleLaw50.FiniteWindow
Source
Little, Little's Law as Viewed on Its 50th Anniversary, Oper. Res. 59(3) (2011), DOI 10.1287/opre.1110.0940, p. 539, §2.2.3, Corollary (3)(b)
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.