Research Topics | 研究テーマ – Giancarlos Troncoso Parady

The effect of new transportation technologies on travel behavior and urban form
Travel behavior changes during the COVID-19 Pandemic
Social networks and travel behavior
Passive methods to detect joint activities in social networks
Validation methods for discrete choice models
Evacuation behavior during natural disasters
Built environment and travel behavior

■　As a research unit we are working on more topics (see our laboratory’s output here)

① The effect of new transportation technologies on travel behavior and urban form

New transportation technologies such as automated vehicles promise to be as disruptive, if not more than the advent of the automobile. In this project we attempt to understand the effect such technologies will have on our travel behavior and our cities in order to minimize its negative impacts while harnessing its potential.

Hansen accessibility changes based on simulation of Private Autonomous Vehicle introduction in Gunma Prefecture. Darker colors show higher increases in accessibility (Luo, Parady, Takami & Harata 2019) .

② Travel behavior changes during the COVID-19 Pandemic

Simulation of the effects of perception of degree of self-restriction of others and COVID-19 dread on going-out self-restriction (“stay home”) choice probability for eating-out and leisure activities. (Parady, Taniguchi & Takami, 2020).

In this project we attempt to clarify how has travel behavior changed during the pandemic and what factors explain the observed changes. In particular we focus on the effects of risk perception and social influence.

③ Social networks and travel behavior

Social network distribution of residents in the Greater Tokyo Area (Parady, Takami & Harata, 2020)

Comparison of mode specific market shares of social interaction by geographical distance (Parady et al., 2021)

Our travel patterns are interdependent with the patterns of the people in our social networks, and while joint activities account a large share of total travel, this is still an understudied issue in the field.

This projects seeks to shed some light on the factors that affect social interaction among social networks and the decision making process behind group activities with the aim of improving the predictive ability of travel demand models.

④ Passive methods to detect joint activities in social networks

Ground truth data plotted against Google Maps Location History data. Part of an experiment to evaluate the effectiveness of using Google Maps Location History data to detect joint activities in social networks (Suzuki, Parady, Oyama & Chikaraishi, 2021)

One of the key barriers in the study of travel behavior in social networks is the extreme response burden of travel behavior and social network surveys. As such, findings passive ways to collect these types of data is one of the upmost importance.

This projects evaluates the effectiveness of different passive data collection methods such as Google Maps Location History Data and GPS data to detect joint activities in social networks.

⑤ Validation methods for discrete choice models

The concept of model validation illustrated (Parady, Ory & Walker, 2021)

Despite a strong dependence on cross-sectional data and the difficulties associated with conducting experiments in the transportation field, only 18.1% of articles in the field report model validation (Parady, Ory & Walker, 2021).

In addition to understand the current state of affairs regarding model reporting and validation practices, this project seeks to evaluate different validation methods and performance indices in order identify the most adequate methods given the type of models and typical datasets in the field.

⑥ Evacuation behavior during natural disasters

(Left) Basic description of target prefectures and municipalities. Data in brackets after municipality name provide further seismic and topographical information. (Right) Plot of evacuation probability considering evacuation warning and seawall height (Parady, Tran & Gilmour, 2019)

Short-notice natural disasters such as tsunamis present immense challenges to planners in terms of disaster preparedness.

This study seeks to understand evacuation patterns during short-notice natural disasters (prompt departure, destination choice, etc.) Such knowledge is expected to contribute to drawing better evacuation planning strategies taking into consideration, among others, psychological factors that might infuse a false sense of security on residents.

⑦ Built environment and travel behavior

Urbanization level of Fukuoka City estimated as a continuous latent variable thus avoiding the arbitrary urban-suburban definition (Parady, Takami & Harata, 2017)

Recent years have seen a paradigm shift in the conceptualization of what constitutes good urban development. Be it New Urbanism Smart Growth, or Compact Cities, one of the main premises behind these new paradigms is that mixed-use, high density developments can reduce automobile dependency and promote alternative modes such as transit, bicycles or walking (Parady, Takami & Harata, 2017).

This project seeks to quantify the magnitude of the built environment effect on travel behavior, while controlling for possible confounding factors such as residential self-selection.