Eqs. (6.1)-(6.2) — EMSR(S) is non-increasing in S
ProvedSeatInventory.Distinct.emsr_antitoneemsrp2o-batch-pfp1ap2o-gran-per-chapterp2o-plan-bookp2o-v1revenue-managementseat-inventory-control
Let a fare class have average fare and integer-valued requests . Then the tail probability is a non-increasing function of , and so is the expected marginal seat revenue
Monotonicity of the marginal values is what makes allocating seats one at a time, always to the largest remaining marginal value, revenue-maximising.
Formalization Note The thesis says "decreasing"; its justification (point probabilities are nonnegative) gives only non-increasing, and strict decrease is false when is bounded, so the statement is Antitone. Nonnegativity of the fare is required for the EMSR part and is a standing convention of the model (a fare is a price received).
Preamble
import Mathlib import Definitions.Def_SeatInventory_Distinct_DemandModel
Formal statement
namespace SeatInventory.Distinct
/-- Belobaba 1987, Eqs. (6.1)–(6.2), p. 142: `P̄(S) = P[r ≥ S]` is non-increasing in `S`, hence so
is `EMSR(S) = P̄(S) · f` for a nonnegative fare `f`. -/
theorem emsr_antitone (f : ℝ) (hf : 0 ≤ f) (p : PMF ℕ) :
Antitone (tailProb p) ∧ Antitone (emsr f p) := by sorry
end SeatInventory.Distinct
Source
Belobaba, Air Travel Demand and Airline Seat Inventory Management, MIT Flight Transportation Laboratory Report R87-7 (PhD thesis), 1987, p. 142, Eqs. (6.1)-(6.2) and the paragraph after them
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.