Calibration of a transit route choice model using revealed population data of smartcard in a multimodal transit network
- PDF / 1,028,160 Bytes
- 24 Pages / 439.37 x 666.142 pts Page_size
- 8 Downloads / 194 Views
Calibration of a transit route choice model using revealed population data of smartcard in a multimodal transit network Ikki Kim1 · Hyoung‑Chul Kim2 · Dong‑Jeong Seo3 · Jung In Kim1
© Springer Science+Business Media, LLC, part of Springer Nature 2019
Abstract One of the major objectives of this study is to provide more realistic and accurate results related to transit passenger’s route choice behavior by using population data of revealed preference from smartcard transaction records. The smartcard data of the Seoul city provides both boarding and alighting location and time, which can make possible to trace each passenger’s actually used path trajectory with close to 100% market penetration of smartcard usage. This study built an abstract transit network with representative nodes by aggregating all near-by bus stops within walkable distance and with abstract paths by aggregating lines for a specific OD pair that run the same trajectory links by same transit modes. This complex and huge-scale transit network allowed to analyze the route choice behavior of transit passengers in a multimodal transit system that could not be found from the data of relatively small-size cities. This study selected OD pairs which had two or more alternative paths in order to analyze choice behavior requiring a plural alternative choice set. The number of the selected OD pairs are 124,393 pairs that are 33.9% of whole OD pairs that has two or more trip records. The calibration result showed that it is good statistically and logically to include the six explanatory variables in the utility function of the multinomial Logit model. Those are in-vehicle travel time, out-of-vehicle travel time, transfer penalty index, travel time reliability measure, and path circuity index. Keywords Route choice behavior · Transit assignment · Smartcard data · Transit fare system · A trajectory of the chosen path
* Ikki Kim [email protected] 1
Department of Transportation and Logistics Engineering, Hanyang University, 55 Hanyang Deahak‑ro, Sangrok‑gu, Ansan, Gyunggi‑do 15588, South Korea
2
ChungNam Institute, Gongju, Chungcheongnam‑do, South Korea
3
Korea Transportation Safety Authority, Gimcheon, Gyeongsangbuk‑do, South Korea
13
Vol.:(0123456789)
Transportation
Introduction A route choice model is an inevitable basic model both for highway assignment and for transit assignment. The model is required to represent realistic traveler’s route choice behavior accurately in order to get realistic flows/passengers in a network as the result of assignment analysis. Better calibration of the models requires ideally revealed and population data of each traveler’s chosen path trajectory and its choice set. However, the collection of exact trajectory data for each traveler’s actually used path has been extremely difficult. As results of such difficulty, many studies have assumed that travelers would choose minimum travel time path mathematically as typically explained in Sheffi (1985) while also considering monetary terms of fare or toll by conver
Data Loading...
