Simulating individual work trips for transit-facilitated accessibility study

Research output: Contribution to journalArticle

Abstract

To measure job accessibility, person-based approaches have the advantage to capture all accessibility components: land use, transportation system, individual’s mobility and travel preference, as well as individual’s space and time constraints. This makes person-based approaches more favorable than traditional aggregated approaches in recent years. However, person-based accessibility measures require detailed individual trip data which are very difficult and expensive to acquire, especially at large scales. In addition, traveling by public transportation is a highly time sensitive activity, which can hardly be handled by traditional accessibility measures. This paper presents an agent-based model for simulating individual work trips in hoping to provide an alternative or supplementary solution to person-based accessibility study. In the model, population is simulated as three levels of agents: census tracts, households, and individual workers. And job opportunities (businesses) are simulated as employer agents. Census tract agents have the ability to generate household and worker agents based on their demographic profiles and a road network. Worker agents are the most active agents that can search jobs and find the best paths for commuting. Employer agents can estimate the number of transit-dependent employees, hire workers, and update vacancies. A case study is conducted in the Milwaukee metropolitan area in Wisconsin. Several person-based accessibility measures are computed based on simulated trips, which disclose low accessibility inner city neighborhoods well covered by a transit network.

Original languageEnglish (US)
JournalEnvironment and Planning B: Urban Analytics and City Science
DOIs
StateAccepted/In press - Jan 1 2017

Fingerprint

accessibility
human being
worker
employer
census
job search
public transportation
road network
transportation system
agglomeration area
commuting
land use
travel
employee
metropolitan area
Land use
Vacancies
ability
Personnel
time

Keywords

  • accessibility
  • agent-based modeling
  • Geographic information systems
  • simulation
  • transit
  • work trip

ASJC Scopus subject areas

  • Geography, Planning and Development
  • Architecture
  • Environmental Science(all)
  • Urban Studies
  • Nature and Landscape Conservation
  • Management, Monitoring, Policy and Law

Cite this

@article{a207e357c1204bad9e9177896c2a7db7,
title = "Simulating individual work trips for transit-facilitated accessibility study",
abstract = "To measure job accessibility, person-based approaches have the advantage to capture all accessibility components: land use, transportation system, individual’s mobility and travel preference, as well as individual’s space and time constraints. This makes person-based approaches more favorable than traditional aggregated approaches in recent years. However, person-based accessibility measures require detailed individual trip data which are very difficult and expensive to acquire, especially at large scales. In addition, traveling by public transportation is a highly time sensitive activity, which can hardly be handled by traditional accessibility measures. This paper presents an agent-based model for simulating individual work trips in hoping to provide an alternative or supplementary solution to person-based accessibility study. In the model, population is simulated as three levels of agents: census tracts, households, and individual workers. And job opportunities (businesses) are simulated as employer agents. Census tract agents have the ability to generate household and worker agents based on their demographic profiles and a road network. Worker agents are the most active agents that can search jobs and find the best paths for commuting. Employer agents can estimate the number of transit-dependent employees, hire workers, and update vacancies. A case study is conducted in the Milwaukee metropolitan area in Wisconsin. Several person-based accessibility measures are computed based on simulated trips, which disclose low accessibility inner city neighborhoods well covered by a transit network.",
keywords = "accessibility, agent-based modeling, Geographic information systems, simulation, transit, work trip",
author = "Ruihong Huang",
year = "2017",
month = "1",
day = "1",
doi = "10.1177/2399808317702148",
language = "English (US)",
journal = "Environment and Planning B: Planning and Design",
issn = "0265-8135",
publisher = "Pion Ltd.",

}

TY - JOUR

T1 - Simulating individual work trips for transit-facilitated accessibility study

AU - Huang, Ruihong

PY - 2017/1/1

Y1 - 2017/1/1

N2 - To measure job accessibility, person-based approaches have the advantage to capture all accessibility components: land use, transportation system, individual’s mobility and travel preference, as well as individual’s space and time constraints. This makes person-based approaches more favorable than traditional aggregated approaches in recent years. However, person-based accessibility measures require detailed individual trip data which are very difficult and expensive to acquire, especially at large scales. In addition, traveling by public transportation is a highly time sensitive activity, which can hardly be handled by traditional accessibility measures. This paper presents an agent-based model for simulating individual work trips in hoping to provide an alternative or supplementary solution to person-based accessibility study. In the model, population is simulated as three levels of agents: census tracts, households, and individual workers. And job opportunities (businesses) are simulated as employer agents. Census tract agents have the ability to generate household and worker agents based on their demographic profiles and a road network. Worker agents are the most active agents that can search jobs and find the best paths for commuting. Employer agents can estimate the number of transit-dependent employees, hire workers, and update vacancies. A case study is conducted in the Milwaukee metropolitan area in Wisconsin. Several person-based accessibility measures are computed based on simulated trips, which disclose low accessibility inner city neighborhoods well covered by a transit network.

AB - To measure job accessibility, person-based approaches have the advantage to capture all accessibility components: land use, transportation system, individual’s mobility and travel preference, as well as individual’s space and time constraints. This makes person-based approaches more favorable than traditional aggregated approaches in recent years. However, person-based accessibility measures require detailed individual trip data which are very difficult and expensive to acquire, especially at large scales. In addition, traveling by public transportation is a highly time sensitive activity, which can hardly be handled by traditional accessibility measures. This paper presents an agent-based model for simulating individual work trips in hoping to provide an alternative or supplementary solution to person-based accessibility study. In the model, population is simulated as three levels of agents: census tracts, households, and individual workers. And job opportunities (businesses) are simulated as employer agents. Census tract agents have the ability to generate household and worker agents based on their demographic profiles and a road network. Worker agents are the most active agents that can search jobs and find the best paths for commuting. Employer agents can estimate the number of transit-dependent employees, hire workers, and update vacancies. A case study is conducted in the Milwaukee metropolitan area in Wisconsin. Several person-based accessibility measures are computed based on simulated trips, which disclose low accessibility inner city neighborhoods well covered by a transit network.

KW - accessibility

KW - agent-based modeling

KW - Geographic information systems

KW - simulation

KW - transit

KW - work trip

UR - http://www.scopus.com/inward/record.url?scp=85041616706&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85041616706&partnerID=8YFLogxK

U2 - 10.1177/2399808317702148

DO - 10.1177/2399808317702148

M3 - Article

JO - Environment and Planning B: Planning and Design

JF - Environment and Planning B: Planning and Design

SN - 0265-8135

ER -