@article{19c5179abaea4da0901be9d55e9944ca,
title = "A new schedule-based transit assignment model with travel strategies and supply uncertainties",
abstract = "This paper proposes a new scheduled-based transit assignment model. Unlike other schedule-based models in the literature, we consider supply uncertainties and assume that users adopt strategies to travel from their origins to their destinations. We present an analytical formulation to ensure that on-board passengers continuing to the next stop have priority and waiting passengers are loaded on a first-come-first-serve basis. We propose an analytical model that captures the stochastic nature of the transit schedules and in-vehicle travel times due to road conditions, incidents, or adverse weather. We adopt a mean variance approach that can consider the covariance of travel time between links in a space-time graph but still lead to a robust transit network loading procedure when optimal strategies are adopted. The proposed model is formulated as a user equilibrium problem and solved by an MSA-type algorithm. Numerical results are reported to show the effects of supply uncertainties on the travel strategies and departure times of passengers.",
keywords = "Schedule-based transit assignment, Strategy, Supply uncertainty, User equilibrium",
author = "Younes Hamdouch and Szeto, {W. Y.} and Y. Jiang",
note = "Funding Information: The work described in this paper was partially supported by Grants from the United Arab Emirates University – National Research Foundation (UAEU-NRF-58), the Central Policy Unit of the Government of the Hong Kong Special Administrative Region and the Research Grants Council of the Hong Kong Special Administrative Region, China (HKU7026-PPR-12), the Hong Kong University Research Committee (201211159009), the National Natural Science Foundation of China (71271183), and a Research Postgraduate Studentship from the University of Hong Kong. Appendix (i) We first show that σ jk 2 = ϕ σ ij 2 + ( 1 - ϕ ) Var ( Y jk ) + ϕ ( 1 - ϕ ) μ ij - E ( Y jk ) 2 σ jk 2 = ∑ t t 2 P ( T jk = t ) - ∑ t tP ( T jk = t ) 2 = ∑ t t 2 ϕ P ( T jk = t ) + ( 1 - ϕ ) P ( Y jk = t ) - ∑ t t ϕ P ( T jk = t ) + ( 1 - ϕ ) P ( Y jk = t ) 2 = ϕ ∑ t t 2 P ( T ij = t ) + ( 1 - ϕ ) ∑ t t 2 P ( Y jk = t ) - ϕ 2 ∑ t tP ( T ij = t ) 2 - ( 1 - ϕ ) 2 ∑ t tP ( Y jk = t ) 2 - 2 ϕ ( 1 - ϕ ) ∑ t tP ( T ij = t ) ∑ t tP ( Y jk = t ) = ϕ E ( T ij 2 ) - ϕ 2 ( E ( T ij ) ) 2 + ( 1 - ϕ ) E ( Y jk 2 ) - ( 1 - ϕ ) 2 ( E ( T jk ) ) 2 - 2 ϕ ( 1 - ϕ ) E ( T ij ) E ( Y jk ) = ϕ E ( T ij 2 ) - ( E ( T ij ) ) 2 + ( 1 - ϕ ) E ( Y jk 2 ) - ( E ( Y jk ) ) 2 + ϕ ( 1 - ϕ ) ( E ( T ij ) ) 2 + ϕ ( 1 - ϕ ) ( E ( Y jk ) ) 2 - 2 ϕ ( 1 - ϕ ) E ( T ij ) E ( Y jk ) = ϕ σ ij 2 + ( 1 - ϕ ) Var ( Y jk ) + ϕ ( 1 - ϕ ) ( E ( T ij ) ) 2 + ( E ( Y jk ) ) 2 - 2 E ( T ij ) E ( Y jk ) = ϕ σ ij 2 + ( 1 - ϕ ) Var ( Y jk ) + ϕ ( 1 - ϕ ) E ( T ij ) - E ( Y jk ) 2 = ϕ σ ij 2 + ( 1 - ϕ ) Var ( Y jk ) + ϕ ( 1 - ϕ ) μ ij - E ( Y jk ) 2 . (ii) For each 1 ⩽ n ⩽ N l - 2 , we will show by induction on n ′ , 1 ⩽ n ′ ⩽ N l - 1 - n that: Cov ( T i n ( l ) i n + 1 ( l ) , T i n + n ′ ( l ) i n + n ′ + 1 ( l ) ) = ϕ n ′ σ i n ( l ) i n + 1 ( l ) 2 . – n ′ = 1 Cov ( T i n ( l ) i n + 1 ( l ) , T i n + 1 ( l ) i n + 2 ( l ) ) = Cov ( T i n ( l ) i n + 1 ( l ) , ϕ T i n ( l ) i n + 1 ( l ) + ( 1 - ϕ ) Y i n + 1 ( l ) i n + 2 ( l ) ) = Cov ( T i n ( l ) i n + 1 ( l ) , ϕ T i n ( l ) i n + 1 ( l ) ) ( since T jk and Y jk are independent ) = E ( ϕ T i n ( l ) i n + 1 ( l ) 2 ) - E ( T i n ( l ) i n + 1 ( l ) ) E ( ϕ T i n ( l ) i n + 1 ( l ) ) = ϕ E ( T i n ( l ) i n + 1 ( l ) 2 ) - ( E ( T i n ( l ) i n + 1 ( l ) ) ) 2 = ϕ σ i n ( l ) i n + 1 ( l ) 2 . – Assume Cov ( T i n ( l ) i n + 1 ( l ) , T i n + n ′ ( l ) i n + n ′ + 1 ( l ) ) = ϕ n ′ σ i n ( l ) i n + 1 ( l ) 2 – n ′ + 1 Cov ( T i n ( l ) i n + 1 ( l ) , T i n + n ′ + 1 ( l ) i n + n ′ + 2 ( l ) ) = Cov ( T i n ( l ) i n + 1 ( l ) , ϕ T i n + n ′ ( l ) i n + n ′ + 1 ( l ) + ( 1 - ϕ ) Y i n + n ′ + 1 ( l ) i n + n ′ + 2 ( l ) ) = Cov ( T i n ( l ) i n + 1 ( l ) , ϕ T i n + n ′ ( l ) i n + n ′ + 1 ( l ) ) = ϕ Cov ( T i n ( l ) i n + 1 ( l ) , T i n + n ′ ( l ) i n + n ′ + 1 ( l ) ) = ϕ ϕ n ′ σ i n ( l ) i n + 1 ( l ) 2 = ϕ n ′ + 1 σ i n ( l ) i n + 1 ( l ) 2 . ",
year = "2014",
month = sep,
doi = "10.1016/j.trb.2014.05.002",
language = "English",
volume = "67",
pages = "35--67",
journal = "Transportation Research, Series B: Methodological",
issn = "0191-2615",
publisher = "Elsevier Limited",
}