The Future Is Not What It Used To Be In Regional Scenario Planning
Corresponding Author: Patrick Hall, Atlanta Regional Commission
Presented By: Patrick Hall, Atlanta Regional Commission
Abstract
In 2016, the Atlanta Regional Commission (ARC), through a SHRP2 grant, developed four socio-demographic and land use scenarios for the year 2050, based on the previous work of Impacts 2050 (part of NCHRP 750. At the earliest stages of the study, the Regional Strategic Planning Model (RSPM) was chosen as the modelling tool, as the ARC’s regional travel demand model is a complex activity based model (ABM), and was ill suited for the potentially hundreds of runs needed to model the various infrastructure and policy alternatives potentially required for this study.
After the initial meetings to determine the “drivers of change” for future scenarios, four input variables were chosen to represent the differences between the four separate scenarios developed from Impacts 2050. These variables have “low,” “medium,” and “high” settings, requiring 324 separate model runs. The four data inputs variables are:
• Arterial roadway growth
• Transit service growth
• Autonomous vehicles and car sharing
• Congestion pricing
An interactive online tool was developed to allow people to choose their own scenario, and change the input variables between the “low,” “medium,” and “high” settings. As the user chooses the levels between the input variables, the output display is updated, based on the previous 324 RSPM runs. Three of the model outputs are also shown per capita, as there are different population totals between the four scenarios.
The five data outputs are:
• Annual average delay
• Vehicle miles traveled per capita
• Transit ridership (also per capita)
• Bicycle and walking trips (also per capita)
• CO2 emissions (also per capita)
With the success of using RSPM for scenario planning in a very large urbanized region, the paradigm of travel demand modelling is shifting away from a “bottoms-up” approach, as has been done in the past. Instead, there is the possibility of combining both a “top-down” and “bottoms-up” approach, using higher-level models such as RSPM to inform the regional modelling data, and collected “big data,” such as travel speed data, person trips, etc. to inform the disaggregated data at the traffic analysis zone.