Separation of car-dependent commuters from normal-choice riders in mode-choice analysis |
| |
Authors: | Keemin Sohn Jiyoung Yun |
| |
Institution: | (1) Metropolitan Planning Research Group, Seoul Development Institute, 391 Seocho-dong Seocho-gu, Seoul, 137-071, Korea |
| |
Abstract: | The ever-increasing use of cars is a big problem in metropolitan areas. To manage the traffic stream and alleviate air pollution,
most metropolitan governments are attempting to discourage the use of cars. Nevertheless, the results have not been satisfactory.
It is well known that normal-choice riders choose their travel mode based on utility, which is determined by mode-specific
impedances and individual characteristics. On the other hand, this study focuses on identifying car-dependent commuters who
tend to keep driving cars regardless of the circumstances they are confronted with. For this study, psychometric factors characterizing
car-dependent commuters were investigated. However, the performance of the mode-choice model was not sufficiently enhanced
despite incorporation of the psychometric factors. The performance improved considerably when the car-dependent commuters
were excluded. Based on psychometric factors, the support vector machine successfully separated the car-dependent commuters
from normal-choice riders.
|
| |
Keywords: | Car-dependence Mode-choice Latent variables Support vector machine (SVM) |
本文献已被 SpringerLink 等数据库收录! |
|