Example 1, Case 1, p. 292 — from (ξ, 0, 0, 0) the clearing policy returns to (λξ + β, 0, 0, 0) at T₁ = (λ + τ₂/(1 − τ₂))ξ + α
ProvedKumarSeidman.Reentrant.case1_cycle_mapConsider the re-entrant line of Example 1 with processing times satisfying the critical condition and the capacity conditions
and with positive set-up times . Let be as in Example 1.
Then there is such that for every with the following holds. Along every clearing trajectory started at with machines 1 and 2 set up for buffers 4 and 3, at the time
one has
and machines 1 and 2 are again set up for buffers 4 and 3.
This is one full cycle of the clearing policy: the system returns to a state of the same shape with buffer 1 multiplied by and increased by , which drives the instability of Example 1.
Formalization Note The paper states the cycle from an arbitrary start time ; it is stated here from , the initial state of Case 1 (). "Set up for" at means the run of the machine that started before and has not been replaced before : at machine 1 is commencing its set-up for buffer 1. The threshold may depend on and (" is large enough").
import Mathlib import Definitions.Def_KumarSeidman_Reentrant_System import Definitions.Def_KumarSeidman_Reentrant_Trajectory import Definitions.Def_KumarSeidman_Reentrant_Clearing import Definitions.Def_KumarSeidman_Reentrant_Example1
namespace KumarSeidman.Reentrant
/-- **Example 1, Case 1: the cycle map** (§III, p. 292). With `τ₁, …, τ₄ > 0` satisfying
(3)–(5) and set-up times `δ₁, …, δ₄ > 0`, there is `ξ₀` such that for every `ξ > 0` with
`ξ ≥ ξ₀` and every clearing trajectory started at `x(0) = (ξ, 0, 0, 0)` with machines 1 and 2
set up for buffers 4 and 3, at the time `T₁ := (λ + τ₂/(1 - τ₂))ξ + α` one has
`x(T₁) = (λξ + β, 0, 0, 0)` and machines 1 and 2 are again set up for buffers 4 and 3, where
`λ = τ₄/(1 - τ₂)`, `α = (τ₄ + 1)(δ₁ + δ₂)/(1 - τ₂) + δ₃(τ₄ + 1) + δ₄` and
`β = τ₄(δ₁ + δ₂)/(1 - τ₂) + τ₄δ₃ + δ₄`. -/
theorem case1_cycle_map (τ : Fin 4 → ℝ) (hτ : ∀ k, 0 < τ k)
(h3 : 1 < τ 1 + τ 3) (h4 : τ 0 + τ 3 < 1) (h5 : τ 1 + τ 2 < 1)
(δ : Fin 4 → ℝ) (hδ : ∀ k, 0 < δ k) :
∃ ξ₀ : ℝ, ∀ ξ : ℝ, 0 < ξ → ξ₀ ≤ ξ →
∀ T : Trajectory (ex1 τ hτ δ (fun k => (hδ k).le)),
IsClearing (ex1 τ hτ δ (fun k => (hδ k).le)) T → Ex1Initial T ξ →
let T₁ := (ex1Lambda τ + τ 1 / (1 - τ 1)) * ξ + ex1Alpha τ δ
T.x (ex1Buf 0) T₁ = ex1Lambda τ * ξ + ex1Beta τ δ ∧
T.x (ex1Buf 1) T₁ = 0 ∧ T.x (ex1Buf 2) T₁ = 0 ∧ T.x (ex1Buf 3) T₁ = 0 ∧
T.IsSetUpFor 0 T₁ (ex1Buf 3) ∧ T.IsSetUpFor 1 T₁ (ex1Buf 2) := by sorry
end KumarSeidman.Reentrant
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.