Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Proposition 8, p. 29 — the monopoly's optimal AV capacity is N* = n*_{HV,K}, K the smallest i with Σ_{K+1}^I P(m_i) < c_f/(w₀ − c_v)

Proved
AVRideHail.MonopolyCapacity.proposition8_optimal_capacity

by mikedeng1 · Oct 14, 2026 · Mathlib 0df444a (Lean v4.33.1)

autonomous-vehiclescapacity-planningmonopolynewsvendorp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-bookp2o-v1ride-hailing

Consider a ride-hailing market in which a monopoly supplier of autonomous vehicles (AVs) and human drivers (HVs) operate on a common platform. The market has pickup-time function t1(s)=as−rt_1(s) = a s^{-r}t1​(s)=as−r with a>0a > 0a>0 and 0<r<10 < r < 10<r<1, trip time t2>0t_2 > 0t2​>0, maximal rider value V>0V > 0V>0, and perfectly elastic HV supply at reservation earnings w0w_0w0​ (α=0\alpha = 0α=0). AVs cost cv>0c_v > 0cv​>0 per operating hour and cf>0c_f > 0cf​>0 per calendar hour, with cv+cf<w0c_v + c_f < w_0cv​+cf​<w0​ (Assumption 1).

Demand is random with I≥1I \ge 1I≥1 scenarios: potential demand masses 0<m1<m2<⋯<mI0 < m_1 < m_2 < \dots < m_I0<m1​<m2​<⋯<mI​ with probabilities P(mi)≥0P(m_i) \ge 0P(mi​)≥0, ∑iP(mi)=1\sum_i P(m_i) = 1∑i​P(mi​)=1. Each scenario is a loose market (condition (33)); dHV,i∗d^*_{HV,i}dHV,i∗​ is its maximal stable pure-HV equilibrium demand and nHV,i∗=n‾(dHV,i∗)n^*_{HV,i} = \underline{n}(d^*_{HV,i})nHV,i∗​=n​(dHV,i∗​) the pure-HV equilibrium fleet size. The supplier first fixes a capacity N≥0N \ge 0N≥0 and then, in each scenario, a fleet nAV,i∈[0,N]n_{AV,i} \in [0, N]nAV,i​∈[0,N], maximizing (5):

max⁡N≥0  ∑i=1IP(mi){max⁡0≤nAV,i≤NπiC(nAV,i) nAV,i}−cfN,\max_{N \ge 0} \; \sum_{i=1}^I P(m_i) \Big\{\max_{0 \le n_{AV,i} \le N} \pi^C_i(n_{AV,i})\, n_{AV,i}\Big\} - c_f N,N≥0max​i=1∑I​P(mi​){0≤nAV,i​≤Nmax​πiC​(nAV,i​)nAV,i​}−cf​N,

where πiC\pi^C_iπiC​ is the common-platform variable profit per AV (251).

Let KKK be the smallest index iii such that ∑j=i+1IP(mj)<cf/(w0−cv)\sum_{j=i+1}^I P(m_j) < c_f/(w_0 - c_v)∑j=i+1I​P(mj​)<cf​/(w0​−cv​). Then

N∗=nHV,K∗N^* = n^*_{HV,K}N∗=nHV,K∗​

is an optimal capacity for (5).

The result says the monopoly's capacity problem is a newsvendor problem: the optimal capacity is the pure-HV fleet size of the scenario at the critical fractile cf/(w0−cv)c_f/(w_0 - c_v)cf​/(w0​−cv​) of the demand distribution.

Formalization Note. Scenarios are indexed by Fin I, from 000 to I−1I - 1I−1; the page's ∑K+1I\sum_{K+1}^I∑K+1I​ is the sum over indices strictly above KKK, and "the smallest iii" is IsLeast. The page's "in a loose market" is the loose-market condition (33) in every scenario. The conclusion is that nHV,K∗n^*_{HV,K}nHV,K∗​ maximizes (5) over N≥0N \ge 0N≥0, not that it is the unique maximizer: when ∑j≥KP(mj)=cf/(w0−cv)\sum_{j \ge K} P(m_j) = c_f/(w_0 - c_v)∑j≥K​P(mj​)=cf​/(w0​−cv​) exactly, the objective takes the same value at nHV,K−1∗n^*_{HV,K-1}nHV,K−1∗​, so the page's "the optimal AV capacity" is one of the optima. The variable profit πiC\pi^C_iπiC​ is the paper's summary (251) of Proposition 6, taken as a definition.

Preamble
import Mathlib
import Definitions.Def_AVRideHail_MonopolyCapacity_Model
Formal statement
namespace AVRideHail.MonopolyCapacity

theorem proposition8_optimal_capacity (P : AVRideHail.PureHV.MarketParams) (ha : 0 < P.a) (hr0 : 0 < P.r)
    (hr1 : P.r < 1) (ht2 : 0 < P.t2) (hV : 0 < P.V) (halpha : P.alpha = 0)
    (cv cf : ℝ) (hcv : 0 < cv) (hcf : 0 < cf) (hA1 : cv + cf < P.w0)
    (I : ℕ) (hI : 0 < I) (prob : Fin I → ℝ) (hprob : ∀ i, 0 ≤ prob i) (hsum : ∑ i, prob i = 1)
    (m : Fin I → ℝ) (hm : StrictMono m) (hmpos : ∀ i, 0 < m i)
    (hloose : ∀ i, AVRideHail.PureHV.LooseMarket P (m i))
    (dstar : Fin I → ℝ) (hdstar : ∀ i, AVRideHail.PureHV.IsMaxStableEq P (m i) (dstar i))
    (K : Fin I) (hK : IsLeast {k : Fin I | ∑ j ∈ Finset.Ioi k, prob j < cf / (P.w0 - cv)} K) :
    IsMaxOn (monoProfit P cv cf prob m dstar) (Set.Ici 0) (AVRideHail.PureHV.nMin P (dstar K)) := by sorry

end AVRideHail.MonopolyCapacity
Source
Lian & van Ryzin, Capturing the Benefits of Autonomous Vehicles in Ride-Hailing: The Role of Market Configuration, SSRN 3716491 (version of 23 August 2023), p. 29, Proposition 8, (21); proof App. E.1.3, pp. 86–88
Human review
  • Endorsed by Shuze Chen · Oct 14, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 14, 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