Developing Static and Dynamic Multimodal Transportation System Models to Estimate Individual Commuter Footprints Using ArcGIS, Google Maps and Here360

According to the U.S. Greenhouse Gas (GHG) Emissions and Sinks (2015), 27% of GHG emissions in the U.S. are produced by the Transportation sector, second only to the Electric Power Industry with 29%. A vast majority of these emissions are associated with single occupant (Drive-alone) automobile trips. While many policies exist for incentivizing alternative transportation modes, especially for commuting trips, only a few tools are available for estimating the potential benefits. Furthermore, the potential benefits are estimated using aggregate models and assumptions of commute behavior and existing transportation systems. Accordingly, to improve the accuracy of these estimates and potential impact of these policies, this research focuses on developing models to estimate footprints of individual commuters using real-world multimodal transportation systems.;Two different multimodal transportation system modeling approaches were developed and compared: 1. static models were developed using ESRI's ArcGIS 10.5 and its Network Analyst extension and Model Builder programming language, and 2. dynamic models were developed by the implementation of Google Maps and Here360 APIs in R Studio. Comparisons of the results of the static and dynamic models were further analyzed under two conditions: free flow and congested traffic. The developed multimodal transportation systems included six travel choices: Drive-alone, Bicycle, Walking, Bicycle-bus, Walking-bus and Carpool. The developed models computed individualized commuter estimates for six travel measures: travel time, distance and cost, and CO2, VOC, and NOX emissions. A real-world case study was developed for commuters in Fresno, CA. A mixture of real-life and synthetic data comprising 218 commuters was used in the study.;Conclusions of this research include valuable insights about merits and limitations of the static and dynamic modeling approaches. Additionally, the results include multiple comparisons between the travel measure estimates of the two modeling approaches using different travel modes, and under the two traffic conditions. Outcomes of this research could prove valuable in transforming the existing approaches for estimating impacts of policies on alternative transportation incentives.