The U.S. Census Bureau projects the number of Americans aged 65 and older to reach 88.5 million in 2050, more than double its population in 2010. This older population accounts for 13.0 percent of total population in 2010, and it is projected to grow to 20.2 percent in 2050. Recent forecasts projected by the Atlanta Regional Commission (ARC) also show rapid growth of older population in metro Atlanta. Persons aged 65 and older in metro Atlanta are forecasted to grow from 8.8 percent in 2010 to 18.8 percent in 2040, a 10 percent increase over the 30 year period. Baby boomers and increasing life expectancy are considered responsible for the increase.

The impact of the aging population on travel demand and mobility is often overlooked in regional travel demand modeling. Travel characteristics of the older population such as travel distance, travel mode, time-of-day of travel, and tour/trip purposes are likely different from those of the Millennials. ARC’s activity-based model (ABM) incorporates extensive age and person-type variables in model formulations and is therefore well-suited to modeling changes in travel demand due to aging.

In this paper, ARC’s ABM is tested using two datasets: one with ARC’s projection of persons by age for forecast years, which reflects significant aging, and another with a future population that preserves the person age structure of year 2010. The population input to the model is synthesized to reflect the difference in the age structure. The test results are compared for key travel measures which include work location choices, daily activity patterns, the number of mandatory and non-mandatory tours, tour/trip destinations, time-of-day choices, tour/trip mode choices, transit riders by age, and demand and congestion indicators for transit and highway modes. The ARC’s ABM visualization tool, ABMVIZ, is used to interactively display performance measures in various visualization types including time use diagram and tree map. Finally, the possible implications of future transportation policy decisions are discussed based on model results.