Theorem 1.12(i) (zero backlog): stationary M/G/1 workload Pollaczek–Khintchine limit
OpenKyprianou.Ch01.mg1StationaryWorkload_zeroThis is the zero-backlog stationary workload convergence of Theorem 1.12(i) from Kyprianou (2006).
Consider an M/G/1 queue driven by a compound Poisson arrival process with rate and service time distribution \mu = \int_0^\infty x , dF(x) > 0\rho = \lambda \mu < 1\eta\eta([0, x]) = \frac{1}{\mu} \int_0^x (1 - F(y)) , dy$.
Starting from an initially empty system ( = 0 converges weakly as \to \infty$ to the Pollaczek--Khintchine distribution:
62059 \lim_{t \to \infty} P(W_t(0) \le x) = (1 - \rho) \sum_{k=0}^\infty \rho^k \eta^{*k}([0, x]) = \mathrm{pkCDF}(\eta, \rho, x) 62059
for all \ge 0$.
Mathematically, by the duality lemma for Lévy processes, the workload (0) = \sup_{0 \le s \le t} (X_s - X_t)\overline{Y}t = \sup{0 \le s \le t} (\sum_{i=1}^{N_s} \xi_i - s), almost surely, where is the all-time maximum whose law is given by the Pollaczek--Khintchine formula.
import Definitions.Def_LeanLevy_Processes_CompoundPoisson import Mathlib import Definitions.Def_Kyprianou_Ch01_RiskQueueCore open MeasureTheory ProbabilityTheory Kyprianou.Ch01 Set Filter open scoped NNReal ENNReal set_option autoImplicit false universe u
theorem Kyprianou.Ch01.mg1StationaryWorkload_zero
{Ω : Type u} [MeasurableSpace Ω]
(P : Measure Ω) [IsProbabilityMeasure P]
(arrival claims : ℕ → Ω → ℝ) (rate : ℝ≥0)
(F : Measure ℝ) [IsProbabilityMeasure F]
(hd : IsCompoundPoissonDriver arrival claims rate F P)
(hrate : 0 < rate)
(mu : ℝ) (hmu : 0 < mu)
(hclaimpos : F (Set.Iic (0 : ℝ)) = 0)
(hclaimint : Integrable (fun x : ℝ => x) F)
(hmean : (∫ x, x ∂F) = mu)
(hlight : (rate : ℝ) * mu < 1)
(eta : Measure ℝ) [IsProbabilityMeasure eta]
(hetaSupport : eta (Set.Iio (0 : ℝ)) = 0)
(heta : ∀ x : ℝ, 0 ≤ x →
(eta (Set.Iic x)).toReal = integratedTailCDF F mu x) :
∀ x : ℝ, 0 ≤ x →
Tendsto (fun t : ℝ≥0 =>
(P {ω | queueWorkload arrival claims 0 t ω ≤ x}).toReal)
atTop (nhds (pkCDF eta ((rate : ℝ) * mu) x)) := by
sorry