Microsimulation of vehicle transactions within a life-oriented integrated urban modeling system

Abstract

This paper presents the findings of microsimulating vehicle transaction decisions as a process of first-time vehicle purchase, acquisition, trade, disposal, and transaction or do nothing decisions. Vehicle transaction decision processes are simulated within an agent-based integrated Transport Land Use and Energy (iTLE) modeling system. The influence of life-stage transitions and interactions of decisions at different domains along the life-time is accommodated as lead and lag events within the micro-behavioral models and computational procedures of iTLE. For example, trade and disposal decisions are found to be positively influenced by residential move as 1-year lead and 2-year lead events respectively. The simulation starts by generating baseline information for 2006 and runs for 15 years from 2007 to 2021 for Halifax, Canada. The validation results suggest that the majority of the predicted vehicle ownership levels lie within a few percentage points from the observed population. For instance, zero vehicle ownership is under-represented by 1% only. In terms of microsimulation results, a higher proportion of zero vehicle ownership households is predicted in the Halifax urban core. Households with a higher vehicle ownership level are predicted to reside towards the suburban areas. For vehicle transaction type, a higher proportion of first vehicle purchase is predicted to include lower income younger population. In 2021, the average vehicle per household member is predicted to be 0.72. Around 75% of the households are predicted to own at least one vehicle per household member. At the neighborhood-level, the median value of average vehicle ownership per household is predicted to increase to 1.41 in 2021 from 1.37 in 2007. In summary, this paper offers interesting insights on the spatio-temporal evolution of vehicle ownership and transaction pattern, which will assist in testing alternative transport and land use policies.