TY - JOUR
T1 - A faster path-based algorithm with Barzilai-Borwein step size for solving stochastic traffic equilibrium models
AU - Du, Muqing
AU - Tan, Heqing
AU - Chen, Anthony
N1 - Funding Information:
This research is supported by the Natural Science Foundation of China (No. 71801079), the Research Grants Council of the Hong Kong Special Administrative Region (No. 15212217), the Research Committee of the Hong Kong Polytechnic University (No. 1-ZVJV), and the Research Institute for Sustainable Urban Development at the Hong Kong Polytechnic University (1-BBWF).
Funding Information:
This research is supported by the Natural Science Foundation of China (No. 71801079 ), the Research Grants Council of the Hong Kong Special Administrative Region (No. 15212217 ), the Research Committee of the Hong Kong Polytechnic University (No. 1-ZVJV), and the Research Institute for Sustainable Urban Development at the Hong Kong Polytechnic University (1-BBWF).
Publisher Copyright:
© 2020 Elsevier B.V.
PY - 2021/5/1
Y1 - 2021/5/1
N2 - Step size determination (also known as line search) is an important component in effective algorithmic development for solving the traffic assignment problem. In this paper, we explore a novel step size determination scheme, the Barzilai-Borwein (BB) step size, and adapt it for solving the stochastic user equilibrium (SUE) problem. The BB step size is a special step size determination scheme incorporated into the gradient method to enhance its computational efficiency. It is motivated by the Newton-type methods, but it does not need to explicitly compute the second-order derivative. We apply the BB step size in a path-based traffic assignment algorithm to solve two well-known SUE models: the multinomial logit (MNL) and cross-nested logit (CNL) SUE models. Numerical experiments are conducted on two real transportation networks to demonstrate the computational efficiency and robustness of the BB step size. The results show that the BB step size outperforms the current step size strategies, i.e., the Armijo rule and the self-regulated averaging scheme.
AB - Step size determination (also known as line search) is an important component in effective algorithmic development for solving the traffic assignment problem. In this paper, we explore a novel step size determination scheme, the Barzilai-Borwein (BB) step size, and adapt it for solving the stochastic user equilibrium (SUE) problem. The BB step size is a special step size determination scheme incorporated into the gradient method to enhance its computational efficiency. It is motivated by the Newton-type methods, but it does not need to explicitly compute the second-order derivative. We apply the BB step size in a path-based traffic assignment algorithm to solve two well-known SUE models: the multinomial logit (MNL) and cross-nested logit (CNL) SUE models. Numerical experiments are conducted on two real transportation networks to demonstrate the computational efficiency and robustness of the BB step size. The results show that the BB step size outperforms the current step size strategies, i.e., the Armijo rule and the self-regulated averaging scheme.
KW - Barzilai-Borwein step size
KW - Cross-nested logit
KW - Path-based traffic assignment algorithm
KW - Stochastic user equilibrium
KW - Transportation
UR - http://www.scopus.com/inward/record.url?scp=85091606779&partnerID=8YFLogxK
U2 - 10.1016/j.ejor.2020.08.058
DO - 10.1016/j.ejor.2020.08.058
M3 - Journal article
AN - SCOPUS:85091606779
SN - 0377-2217
VL - 290
SP - 982
EP - 999
JO - European Journal of Operational Research
JF - European Journal of Operational Research
IS - 3
ER -