Latent-segmentation-based approach to investigating spatial transferability of activity-travel models

Zeina Wafa, Chandra R. Bhat, Ram Pendyala, Venu M. Garikapati

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Spatial transferability of travel demand models has been an issue of considerable interest, particularly for small- and medium-sized planning areas that often do not have the resources and staff time to collect large-scale travel survey data and estimate model components native to the region. Traditional approaches to identifying geographic contexts that may borrow and transfer models between one another involve the exogenous a priori identification of a set of variables used to characterize the similarity between geographic regions. However, this ad hoc procedure presents considerable challenges because it is difficult to identify the most appropriate criteria a priori. To address that issue, this paper proposes a latent segmentation approach: the most appropriate criteria for identifying areas with similar profiles are determined endogenously in the model estimation phase. The end products are a set of optimal criteria for clustering regions as well as a fully transferred model, segmented to account for heterogeneity in the population. The method is demonstrated, and its efficacy established through a case study in this paper that uses the National Household Travel Survey data set for information on weekday activities of nonworkers in nine regions in the states of California and Florida. The estimated model is then applied to a context withheld from the original estimation to assess its performance. The method is found to offer a robust mechanism for identifying latent segments and establishing criteria for transferring models between areas.

Original languageEnglish (US)
Pages (from-to)136-144
Number of pages9
JournalTransportation Research Record
Volume2493
DOIs
StatePublished - 2015
Externally publishedYes

Fingerprint

Planning

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Mechanical Engineering

Cite this

Latent-segmentation-based approach to investigating spatial transferability of activity-travel models. / Wafa, Zeina; Bhat, Chandra R.; Pendyala, Ram; Garikapati, Venu M.

In: Transportation Research Record, Vol. 2493, 2015, p. 136-144.

Research output: Contribution to journalArticle

@article{9a44fc48c59548819ce91b330f52ea00,
title = "Latent-segmentation-based approach to investigating spatial transferability of activity-travel models",
abstract = "Spatial transferability of travel demand models has been an issue of considerable interest, particularly for small- and medium-sized planning areas that often do not have the resources and staff time to collect large-scale travel survey data and estimate model components native to the region. Traditional approaches to identifying geographic contexts that may borrow and transfer models between one another involve the exogenous a priori identification of a set of variables used to characterize the similarity between geographic regions. However, this ad hoc procedure presents considerable challenges because it is difficult to identify the most appropriate criteria a priori. To address that issue, this paper proposes a latent segmentation approach: the most appropriate criteria for identifying areas with similar profiles are determined endogenously in the model estimation phase. The end products are a set of optimal criteria for clustering regions as well as a fully transferred model, segmented to account for heterogeneity in the population. The method is demonstrated, and its efficacy established through a case study in this paper that uses the National Household Travel Survey data set for information on weekday activities of nonworkers in nine regions in the states of California and Florida. The estimated model is then applied to a context withheld from the original estimation to assess its performance. The method is found to offer a robust mechanism for identifying latent segments and establishing criteria for transferring models between areas.",
author = "Zeina Wafa and Bhat, {Chandra R.} and Ram Pendyala and Garikapati, {Venu M.}",
year = "2015",
doi = "10.3141/2493-15",
language = "English (US)",
volume = "2493",
pages = "136--144",
journal = "Transportation Research Record",
issn = "0361-1981",
publisher = "US National Research Council",

}

TY - JOUR

T1 - Latent-segmentation-based approach to investigating spatial transferability of activity-travel models

AU - Wafa, Zeina

AU - Bhat, Chandra R.

AU - Pendyala, Ram

AU - Garikapati, Venu M.

PY - 2015

Y1 - 2015

N2 - Spatial transferability of travel demand models has been an issue of considerable interest, particularly for small- and medium-sized planning areas that often do not have the resources and staff time to collect large-scale travel survey data and estimate model components native to the region. Traditional approaches to identifying geographic contexts that may borrow and transfer models between one another involve the exogenous a priori identification of a set of variables used to characterize the similarity between geographic regions. However, this ad hoc procedure presents considerable challenges because it is difficult to identify the most appropriate criteria a priori. To address that issue, this paper proposes a latent segmentation approach: the most appropriate criteria for identifying areas with similar profiles are determined endogenously in the model estimation phase. The end products are a set of optimal criteria for clustering regions as well as a fully transferred model, segmented to account for heterogeneity in the population. The method is demonstrated, and its efficacy established through a case study in this paper that uses the National Household Travel Survey data set for information on weekday activities of nonworkers in nine regions in the states of California and Florida. The estimated model is then applied to a context withheld from the original estimation to assess its performance. The method is found to offer a robust mechanism for identifying latent segments and establishing criteria for transferring models between areas.

AB - Spatial transferability of travel demand models has been an issue of considerable interest, particularly for small- and medium-sized planning areas that often do not have the resources and staff time to collect large-scale travel survey data and estimate model components native to the region. Traditional approaches to identifying geographic contexts that may borrow and transfer models between one another involve the exogenous a priori identification of a set of variables used to characterize the similarity between geographic regions. However, this ad hoc procedure presents considerable challenges because it is difficult to identify the most appropriate criteria a priori. To address that issue, this paper proposes a latent segmentation approach: the most appropriate criteria for identifying areas with similar profiles are determined endogenously in the model estimation phase. The end products are a set of optimal criteria for clustering regions as well as a fully transferred model, segmented to account for heterogeneity in the population. The method is demonstrated, and its efficacy established through a case study in this paper that uses the National Household Travel Survey data set for information on weekday activities of nonworkers in nine regions in the states of California and Florida. The estimated model is then applied to a context withheld from the original estimation to assess its performance. The method is found to offer a robust mechanism for identifying latent segments and establishing criteria for transferring models between areas.

UR - http://www.scopus.com/inward/record.url?scp=84960173483&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84960173483&partnerID=8YFLogxK

U2 - 10.3141/2493-15

DO - 10.3141/2493-15

M3 - Article

VL - 2493

SP - 136

EP - 144

JO - Transportation Research Record

JF - Transportation Research Record

SN - 0361-1981

ER -