index.Rmd

---
always_allow_html: yes
title: |
  | Consistent Mode Choices Across 
  | Multiple Model Frameworks
# Your thesis title should be inverted pyramid-shaped, with the longest
# line on top and shorter lines below. Add a new line as tab-|-space-Words
author: 'Christopher Day'
year: '2022'
department: 'Department of Civil and Construction Engineering'
degree: 'Master of Science'
doctype: 'thesis' # thesis or dissertation
committeeMembers: |
  Gregory S. Macfarlane, Chair 
  
  Grant G. Schultz
  
  Gustavious P. Williams
keywords: 'four step model, discrete choice model, activity-based model, microsimulation tool, mode choice model, tour purpose, multinomial logit, latent class choice model'

address: 
  - code: BYU
    address: Brigham Young University, Civil and Construction Engineering Department, 430 EB, Provo, Utah 84602
  - code: BYUPH
    address: Brigham Young University, Public Health Department, 4103 LSB, Provo, Utah 84602
footnote:
  - code: 1
    text: "Corresponding Author"
date: "`r Sys.Date()`"

# Remove the hashtag to specify which version of output you would like.
# Can only choose one at a time.
site: bookdown::bookdown_site

journal: "submitted to Health and Place"

# Note that a tab is needed on the line after the `|`.
abstract: |
  | Every day individuals make a decision about which modes of transportation they should use. Predicting this behavior perfectly is impossible, however, there are  ways to appoximating mode choice in transportation modeling. In this research we aim to increase the accuracy in predicting mode choice using transportation modeling software. Specifically we determine the significance of creating a consistent mode choice model between an activity-based model and a microsimulation tool. Oftentimes, the outputs of an activity-based model serve as the inputs to a microsimulation tool. Yet, the mode choice models betweeen these two software often vary significantly. Using ActivitySim as the activity-based model and BEAM as the microsimulation tool, we establish a consistent mode choice model within BEAM. We then model the mode choice decisions for agents in the Salt Lake City, Utah region. Their modal distributions and mode choice structures are compared to determine the effect of mode choice consistency. Interestingly, we find that a model that uses a consistent mode choice creates a modal distribution that aligns more closely with target shares. We also find that the introduction of path, person, and location variables in the mode choice utility equation helps better predict mode choice decisions. Further research is necessary in order to understand the effects of identical mode choice models between activity-based models and microsimulation tools.

  
acknowledgments: |
  Students should acknowledge funding sources. They may also use the
  acknowledgment page to express appreciation for the committee members, friends
  or family who provided assistance in research, writing or technical aspects of
  the dissertation, thesis or selected project. Acknowledgements should be simple
  and in good taste.

# use option 'oneside' for ETD
# change this to option 'twoside' for printing
#layout: '12pt, oneside, openright' #thesis
layout: "3p, authoryear, review" #elsevier
# Specify the location of the bibliography below

bibliography: [book.bib] 

# Download your specific csl file and refer to it in the line below.
# csl styles: https://www.zotero.org/styles
csl: elsevier-harvard-italics.csl #elsevier
link-citations: yes

lot: true
lof: true
---