Lead Presenter Biography

Daehyun has more than 15 years of experience in the areas of travel demand modeling and transportation planning. He has M.S. and Ph.D. in the area of Transportation System Engineering in School of Civil, Environmental and Sustainable Engineering at Arizona State University. He had worked as a Postdoctoral Fellow at Georgia Institute of Technology in Atlanta, Georgia. Since 2016, he have worked for Maricopa Association of Governments (MAG). At MAG, he works on improvement of MAG ABM and Tour-base Truck Model. He works sample expansion weights for Household Travel Survey.

Co-Authors

Presentation Description

This paper combined several interrelated updates and improvements to incorporate car type including such characteristics as body type and fuel type in the Maricopa Association of Governments (MAG) Activity-Based Model (ABM). Car type is currently missing as a variable in most ABMs in practice. However, it represents an important variable that affects travel choices since different car types may have different travel cost and even different travel times when certain policies such as preferential use of managed lanes by environmentally clean vehicles, are considered. Furthermore, car type may significantly affect the emissions, and for this reason, a differentiation of traffic flows by car type is a very welcome feature for the Regional Transportation Plan (RTP) / air quality conformity analysis. For this reason, car type should be incorporated in the ABM. However, an introduction of car type as an additional travel dimension is not a trivial extension of the ABM.
This project was devoted to the substantiation of the methodology, identification of the technical details of the ABM sub-models affected by car type, corresponding revisions to the ABM software, and analysis of the results. Addressing vehicle fleet composition in the advanced ABM framework means that all related individual choices starting from car ownership should incorporate car type as an additional dimension. Car differentiation by type affects several ABM sub-models and creates additional linkages between them. The following main structural changes were implemented as part of the current project:

● Revised household car ownership model that reflects vehicle type,
● Special new model for intrahousehold car allocation to trips,
● Revised highway assignment and Level-of-Service (LOS) skimming procedures to support a more detailed calculation of accessibilities - and to provide detailed vehicle-type-specific volumes in the final highway assignment,
● Revised process for building taxi/TNC trip tables (loaded and empty) for assignment that reflects assumptions about their fleet composition.

The introduced changes in the model structure to incorporate vehicle type choice required a systematic restructuring of model inputs. It was recognized that aspects of vehicle type choices such as Electric Vehicles (EVs) and/or small-size cars which are negligible percentages today may become dominant in the future and subject to strong regulatory policies. For this reason, it was important to ensure that multiple scenarios can be conveniently and consistently specified.
Adding details to several sub-models also entailed a round of overall model validation and recalibration, starting from the car ownership model itself but eventually including mode choice, car allocation to trips, highway assignments, etc. The entire ABM is now more sensitive to the input assumptions (such as vehicle operating cost differentiated by vehicle type).
The project also included assumptions based on current policies and anticipated growth of EVs. Detailed analysis of future scenarios was performed since it is expected that the vehicle fleet composition may change significantly. As an example, future scenarios with a high penetration rate of EVs, correspondent assumptions regarding the vehicle operating cost, and their impacts on travel patterns, were analyzed in detail. 

Presentation File