Comparative analysis of transit assignment: evidence from urban railway system in the Tokyo Metropolitan Area |
| |
Authors: | Hironori Kato Yuichiro Kaneko Masashi Inoue |
| |
Institution: | (1) Department of Civil Engineering, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku Tokyo, 113-8656, Japan;(2) Department of Civil Engineering, College of Science and Technology, Nihon University, 1-8-14 Kanda-Surugadai, Chiyoda-ku Tokyo, 101-8308, Japan;(3) Transportation Planning Department, Creative Research and Planning Co., Ltd., 1-5-4 Higashiyama, Meguro-ku Tokyo, 153-0043, Japan |
| |
Abstract: | This paper empirically compares the performance of six traffic assignment methods using the same empirical dataset of route
choice. Multinomial logit (MNL), structured multinomial probit (SMNP), user equilibrium (UE), logit-based stochastic user
equilibrium (SUE), probit-based SUE, and all-or-nothing (AON) assignment methods are applied to the comparative analysis.
The investigated methods include those with three types of error components in their cost functions and two types of flow
dependencies. Four methods of generating the route choice set are also compared for use as stochastic traffic assignment methods.
The revealed preference data of urban rail route choice in the Tokyo Metropolitan Area are used for the case analysis. The
empirical case analysis shows that probit-based SUE provides the best accuracy but requires the longest computation time.
It also shows that the heuristics used to generate the choice set influence the method’s accuracy, while the incorporation
of route commonality and in-vehicle congestion significantly improves its accuracy. Finally, the implications for practical
rail planning are discussed on the basis of the analysis results. |
| |
Keywords: | |
本文献已被 SpringerLink 等数据库收录! |
|