Modeling Daily Travel Choices in an Activity- based Framework considering Spatiotemporal Constraints

Abstract

Abstract Activity-based models (ACBMs) are developed to estimate individual travel patterns and improve the accuracy of forecasting aggregated travel demand. Two essential elements of ACBM are the decision to travel, often known as mobility, and the travel choice purpose of the journey. This study aims to develop sequential models of daily mobility and trip purpose choice using an activity-based framework that incorporates some useful features of the activity-based perspective. Random utility maximization (RUM)-based mobility and trip purpose choice models explain key elements of individuals’ daily activity patterns, incorporating spatiotemporal and socio-demographic characteristics. Based on the national household travel survey data collected in South Korea in 2016, these two models were estimated to represent 70% of the full sample, and applied to the remaining 30%. The mobility model reveals that licensed drivers and high-income travelers are more likely to travel, whereas the elderly and homemakers are less likely to do so. The previously selected trip purpose as well as the total number of trips for each purpose during the departure time of the trip influence the choice of the current trip. Demand forecasting and the policy implications of the empirical results are discussed.