Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

§3, p. 6 — fewer than (n+1)m/ε(n+1)^{m/\varepsilon}(n+1)m/ε schedules of long assignments

Proved
UnrelatedSched.FixedMachines.long_assignment_count

by mikedeng1 · Oct 6, 2026 · Mathlib 0df444a (Lean v4.33.1)

approximation-algorithmscountingp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1scheduling

Let P=(pij)P=(p_{ij})P=(pij​) be an m×nm\times nm×n matrix of positive integer processing times with m≥1m\ge 1m≥1 machines and n≥1n\ge 1n≥1 jobs, let d≥0d\ge 0d≥0 be an integer deadline and ε>0\varepsilon>0ε>0. Call an assignment of job jjj to machine iii long if pij>εdp_{ij}>\varepsilon dpij​>εd, and call a partial schedule of long assignments admissible if the long assignments to each machine take at most ddd time in total. Then

  1. in every admissible schedule of long assignments, every machine carries fewer than 1/ε1/\varepsilon1/ε long assignments; and
  2. the number of admissible schedules of long assignments satisfies
#{L:L admissible}<(n+1)m/ε.\#\{L : L \text{ admissible}\} < (n+1)^{m/\varepsilon}.#{L:L admissible}<(n+1)m/ε.

The paper's words: "No machine can handle 1/ε1/\varepsilon1/ε or more long assignments before time ddd. Thus, for any instance there are less than (n+1)m/ε(n+1)^{m/\varepsilon}(n+1)m/ε schedules of long assignments." The count is what bounds the number of linear programs the procedure AεA_\varepsilonAε​ solves, and hence its running time for fixed mmm and ε\varepsilonε.

Formalization Note The exponent m/εm/\varepsilonm/ε is a real number and the power is the real power Real.rpow; it is not rounded. The hypotheses n≥1n\ge 1n≥1 and m≥1m\ge 1m≥1 are needed for the strict inequality: for n=0n=0n=0 there is exactly one (empty) schedule and (0+1)m/ε=1(0+1)^{m/\varepsilon}=1(0+1)m/ε=1. The count is Set.ncard of a subset of a finite type, so it is the honest cardinality.

Preamble
import Mathlib
import Definitions.Def_UnrelatedSched_FixedMachines_LongAssignment
Formal statement
namespace UnrelatedSched.FixedMachines

/-- §3, p. 6: in an admissible schedule of long assignments for `(P, d)` no machine carries
`1/ε` or more long assignments, and (for `n ≥ 1`, `m ≥ 1`) there are fewer than
`(n + 1)^{m/ε}` admissible schedules of long assignments. -/
theorem long_assignment_count {m n : ℕ} (P : Matrix (Fin m) (Fin n) ℕ)
    (hP : ∀ i j, 0 < P i j) (ε : ℝ) (hε : 0 < ε) (d : ℕ) (hn : 1 ≤ n) (hm : 1 ≤ m) :
    (∀ L, IsAdmissible P ε d L →
      ∀ i, ((Finset.univ.filter (fun j => L j = some i)).card : ℝ) < 1 / ε) ∧
    (({L : Fin n → Option (Fin m) | IsAdmissible P ε d L}.ncard : ℕ) : ℝ) <
      ((n : ℝ) + 1) ^ ((m : ℝ) / ε) := by sorry

end UnrelatedSched.FixedMachines
Source
Lenstra, Shmoys, Tardos, Approximation algorithms for scheduling unrelated parallel machines, CWI Report OS-R8714 (1987), p. 6, Section 3, proof of Theorem 3
Human review
  • Endorsed by Shuze Chen · Oct 6, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 6, 2026

    Confirmed by the mission captain (proposal self-audit).

View graph

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions, with reuse governed by our licensing terms.

How Prove2Me worksResearch paperProve2Me Team
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me