Development of a Household Vehicle Ownership and Fleet Composition Model On-call Consulting Services for Transportation Model Re-calibration

Project: Research project

Project Details


Development of a Household Vehicle Ownership and Fleet Composition Model On-call Consulting Services for Transportation Model Re-calibration DEVELOPMENT OF A HOUSEHOLD VEHICLE FLEET COMPOSITION SIMULATOR PHASE 1 ? SYSTEM DESIGN AND MODEL ESTIMATION The project team will implement a robust management plan to ensure successful completion of the project on-budget and on-time. The team will be led by Professor Ram Pendyala of Arizona State University, who will serve as Principal Investigator of the project. He will be assisted by a highly qualified graduate research assistant, Karthik Konduri, who has extensive experience in survey data analysis, econometric and statistical modeling, and microsimulation modeling of activity-travel patterns of households and persons. Professor Chandra Bhat of the University of Texas at Austin, and his graduate research student, will also be involved in the effort. Dr. Bhat and his students have been instrumental in the development of the MDCEV model and the design of the multiple choice occasion based approach to modeling household vehicle fleet composition and utilization. Over the past few years, Dr. Bhat and his students have developed algorithms and program codes to estimate and test alternative specifications of MDCEV models. More recently, he has developed a mechanism for conveniently and efficiently applying the MDCEV model in a forecasting mode. Professor Bhat and his students will lead the MDCEV model estimation tasks of this project and assist on a variety of other tasks including model calibration/validation and model sensitivity testing and scenario analysis. A unique data set with vehicle type choice and transactions information collected from a large sample of 6000+ households residing in the Southern California region is available to the research team, and will be utilized for this project (in addition to the standard National Household Travel Survey data sets). The Southern California survey data set was collected as part of a survey conducted by the California Energy Commission (CEC). The data set includes both revealed and stated preference information, as well as a variety of dynamic transactions data over time, thus providing a rich source of information for estimating both vehicle type choice and vehicle evolution models. Professor Ram Pendyala at Arizona State University will be responsible for the timely delivery of all project deliverables. He will serve as the central point of contact for all aspects related to this project; however, MAG staff will have the complete flexibility to directly contact and communicate with either of the two entities at any time. This is a team effort with all team members committed to the success of the project. The team members have a long history of working together on projects of this nature and have consistently shown a spirit of flexibility and adaptability necessary to deliver high quality products on time and on budget. Quarterly progress reports will be submitted in conjunction with each quarterly invoice, documenting progress made, work planned for the next quarter, and expenditures to date. In addition to the progress reports, the team will submit one interim technical report after Task 5. At the end of the project, the team will deliver a project final report with detailed model specifications, results of calibration, validation, and sensitivity analysis exercises, and comprehensive descriptive statistics on vehicle ownership, fleet composition, and evolution behavior. The report will also describe how the model system can be implemented within the context of the activity-based microsimulation model system that is being developed for MAG. Clean and documented data sets will be furnished to MAG in an electronic format.
Effective start/end date7/1/125/31/14


  • LOCAL: Arizona Municipal Government: $77,997.00


Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.