Microsimulation of Human Daily Activities Integrating Social Media and Survey Data

Authors: Nana LUO*, San Diego State University, Atsushi Nara, San Diego State University
Topics: Behavioral Geography, Geographic Information Science and Systems, Quantitative Methods
Keywords: Micro-simulation, Activity-based Model, Individual Daily Travel Patterns, Bayesian Network, Gravity Model
Session Type: Paper
Day: 4/13/2018
Start / End Time: 3:20 PM / 5:00 PM
Room: Bayside A, Sheraton, 4th Floor
Presentation File: No File Uploaded

Understanding individual daily activity patterns that are derived from human desire and need to participate in activities has received growing attentions due to its important role in travel demand forecast and urban planning, disaster evacuation and disease spatial transmission. Conventional methods utilize travel diary surveys to characterize individual activities, which have been increasingly criticized because of highly cost, time consuming and low update frequency. In recent decades, social media data has emerged as computer science and online technique developed, and has been widely used in behavior research. In this study, an activity-based model (ABM) using travel survey and census data is built in an individual level: (1) population synthesis sub-model; (2) employment\activity location choice sub-model; (3) activity type, count and transport mode sub-model; and (4) activity duration sub-model. Further, the built location choice model is improved by combining social media and socioeconomic data, and therefore offsets the temporal issue of survey data. And most importantly, the real-time media data can update the model frequently and infer recent individual activity patterns. Finally, a detailed discussion is provided about how different spatial scales influence the simulation results of the built ABM model. This study also demonstrates the potential value of integrating social media and survey data to improve modeling individual daily activities.

Abstract Information

This abstract is already part of a session. View the session here.

To access contact information login