(4.8), p. 10 — f_{i+1} ≥ max[f_1 + t̄_1, …, f_i + t̄_i]
ProvedFulkersonPERT.Bounds.f_ge_max_add_meancritical-pathjensenp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1pert
Let be a project network with events and let bundle distributions be given, each a probability distribution on its finite support. Let be Fulkerson's numbers (4.2) and the expected length of arc . Then for every event ,
This is the inequality obtained by interchanging summation and maximum in (4.6); it is the first half of the induction step of (4.4).
Formalization Note Fulkerson's node is the event , and his arcs are the arcs that exist; the maximum is over those. Arc lengths may be negative (footnote, p. 4).
Preamble
import Mathlib import Definitions.Def_CriticalPath_Events_ProjectNetwork import Definitions.Def_CriticalPath_Events_EventTimes import Definitions.Def_FulkersonPERT_Bounds_Model
Formal statement
namespace FulkersonPERT.Bounds
open CriticalPath.Events
theorem f_ge_max_add_mean {n : ℕ} (N : ProjectNetwork n) (D : BundleDist n) (hD : D.IsProb)
(j : Fin (n + 1)) (hj : j ≠ 0) :
(N.pred j).sup' (N.pred_nonempty hj) (fun i => fNum N D i + meanLength D i j) ≤
fNum N D j := by sorry
end FulkersonPERT.Bounds
Source
Fulkerson, Expected Critical Path Lengths in PERT Networks, RAND Memorandum RM-3075-PR (1962), p. 10, (4.6)–(4.8)
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.