Workshop on Activity-Based Modeling of Spatial and Temporal Patterns of Human Travel Behaviour

July 9-10, 2016: Tongji University, Shanghai, China

CTR will be helping present a new course on activity-based modeling this summer in China. Download the short course agenda (PDF).  Registrants, if you have not already, please fill out this survey.

  • Update as of September 6, 2016: Course materials are now available. Instructions to access and download the course materials have been sent via e-mail to all workshop registrants. Please follow the instructions to access the materials, and contact the course organizers if you have any questions (you can email Maureen Kelly at

This course will introduce attendees to the activity-based approach to travel behavior analysis, focusing on data-supported transportation planning and operations (D-STOP) opportunities, and a number of unique elements of the activity-based modeling framework that have been implemented in the Los Angeles area and are scheduled to be implemented in the New York area. The unified modeling framework, developed at the Center for Transportation Research (CTR) at The University of Texas at Austin, the Georgia Institute of Technology, the University of California at Santa Barbara, and many other universities worldwide, constitutes a pragmatic and field-tested comprehensive approach that is deeply embedded in the foundations of human behavioral decision-making. The model system, which has been extensively tested and applied in a number of settings, advances the state of the art of activity-travel demand forecasting beyond the daily activity-pattern and tour-based approaches employed in several metropolitan areas. It is a next-generation application platform that offers a more flexible solution to contemporary transportation policy analysis, offers a stronger foundation for meeting current and future model requirements, and can take advantage of new technologies and methods as they emerge.

In short, the course will enable researchers and practitioners alike to move to the forefront of model development to address the increasingly complex land-use, built environment, transport, and environmental policies, the analyses of which far exceed the capabilities of the usual modeling/simulation techniques. Download the flyer with complete details (PDF).

Who should attend?

The course is designed for researchers interested in learning about cutting edge travel modeling methods founded in strong behavioural theories, and practitioners (from consulting, local, state, and federal planning agencies, and transit agencies) interested in learning about a flexible and practical suite of modeling tools that may be used to analyse the impacts of a variety of technology, pricing, and land-use policies to better plan, invest, design, and manage transportation systems. Attendees are expected to have a basic knowledge of transportation analysis methods and mathematical/statistical modeling techniques.


The workshop directors are Professors Chandra R. Bhat (CTR, University of Texas at Austin), Ram M. Pendyala (Georgia Tech), and Konstadinos G. Goulias (University of California at Santa Barbara).  The course is hosted by Professor Xin Ye (Tongji University). In addition, expert guest instructors may be invited to present specific workshop modules.  All of the instructors have extensive experience in agent-based modeling of spatial and temporal patterns of human movements, and have been pioneers in the development of concepts and methods that are now widely adopted around the world.

Location and course fees; registration

The course will be hosted by the College of Transportation Engineering of Tongji University, Shanghai, which is one of the key colleges of Tongji University and is recognized worldwide for its leadership and scholarship in the field.  The course is scheduled to commence at 1:00 PM on Saturday, July 9, 2016 and conclude at 4:00 PM on Sunday, July 10, 2016.  Course fees are as follows:

These rates valid until June 10; after June 10, each category increases by $75. The course fee includes access to the complete set of notes and presentations, and program code for model estimation and application.  Attendees are responsible for making their own arrangements for travel and accommodation.  The course organizers will provide light refreshments during morning and afternoon breaks. Lunch is provided on Day 2. (Note that cancellations received by June 15 will be refunded in full but are subject to a $75 cancellation fee. No refunds are offered after June 15.)

Workshop location

The workshop location is the Zonghe Building, Room 405, on the main Campus of Tongji University (1239 Siping Road, Shanghai 200092). (View map.) Here’s a list of nearby hotels:

Hotel Name Price (USD/night) Distance to campus (m) Address
Kingswell Hotel Shanghai $93 350m 50 Zhangwu Road
Jinjiang MetroPolo Hotel Shanghai Tongji University $62 400m 1251 Siping Road
Jinjiang Inn Shanghai Tongji University $38 450m 1251 Siping Road
Tongji Guest House $31 600m 69 Zhangwu Road


Other hotels are available a little further away:

Hotel Name Price (USD/night) Distance to campus (miles) Address
Holiday Inn Express Wujiaochang $50 0.98mi 1729 Huang Xing Road
Guangdong Hotel Shanghai $81 1.66mi 328 Yi Xian Road
Howard Johnson Caida Plaza Shanghai $76 1.73mi 188 Wudong Road Yangpu
Baolong Hotel Shanghai $80 1.89mi 180 Yixian Road

Workshop Logistics and Course Content

Module 1:  Essentials of Activity-Travel Microsimulation (Day 1)

  • Discrete choice models of activity-travel behavior – emerging approaches and behavioral paradigms
  • Population synthesis methods and the PopGen (Population Generation) software
  • Base year demographic models and demographic evolution models
  • Spatio-temporal accessibility computations

Module 2: Behavioral Models – Estimation and Application (Day 2)

  • Activity-based analysis of sense of place and space
  • Understanding time-space geography, interactions, and constraints
  • Activity-travel scheduling and the CEMDAP (Comprehensive Econometric Microsimulator for Daily Activity-travel Patterns) software
  • Estimation of multiple discrete continuous extreme value (MDCEV) models
  • Forecasting with the MDCEV model – Applications to intra-household interactions and household vehicle fleet composition and utilization
  • Sensitivity analysis and model validation
  • Integration of activity-based models with dynamic network models



Posted by Maureen Kelly  |  Category : D-STOP