Corollary 3.2 — completion times of the -schedule via -points
ProvedSingleMachineSched.AlphaSched.corollary_3_2alpha-pointslist-schedulingp2o-batch-p200bp2o-gran-per-chapterp2o-plan-paperp2o-v1scheduling
Let jobs have integral processing times , integral release dates and weights , indexed by nonincreasing . Let with for all , and let be the completion time of job in the -schedule. With the -point of and the fraction of job processed by time , both in the LP schedule,
The sum ranges over all jobs , including , for which and the term is . This is the basic bound from which the paper's analyses of the -schedule and the -schedule start.
Preamble
import Mathlib import Definitions.Def_SingleMachineSched_AlphaSched_AlphaSchedule
Formal statement
namespace SingleMachineSched.AlphaSched
open Classical in
/-- Corollary 3.2 (Goemans et al. 2002, p. 179): for every `α ∈ (0, 1]ⁿ`, the completion time of
job `j` in the `(α_j)`-schedule satisfies
`C^α_j ≤ t_j(α_j) + Σ_{k : α_k ≤ η_k(α_j)} (1 + α_k - η_k(α_j)) p_k`,
where `α`-points and `η` are taken in the LP schedule. The sum includes `k = j`. -/
theorem corollary_3_2 {n : ℕ} (p r : Fin n → ℕ) (w : Fin n → ℝ)
(hp : ∀ j, 0 < p j) (hw : ∀ j, 0 < w j)
(hsort : ∀ j k : Fin n, j ≤ k → w k / p k ≤ w j / p j)
(α : Fin n → ℝ) (hα : ∀ k, 0 < α k ∧ α k ≤ 1) (j : Fin n) :
alphaCompletion p r α j ≤
alphaPoint p (lpSet p r) j (α j) +
∑ k ∈ Finset.univ.filter (fun k => α k ≤ eta p (lpSet p r) j (α j) k),
(1 + α k - eta p (lpSet p r) j (α j) k) * (p k : ℝ) := by sorry
end SingleMachineSched.AlphaSched
Source
Goemans, Queyranne, Schulz, Skutella & Wang, Single Machine Scheduling with Release Dates, SIAM J. Discrete Math. 15(2) (2002), p. 179, Corollary 3.2
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.