The transportation sector is transforming as new technologies, integrate or replace existing systems. Designing transportation solutions for realworld urban scale systems has previously been accomplished with limited analytics because of the computational scale. New large scale computational capabilities (e.g. cloud computing and supercomputing), data analytics (e.g machine learning and intelligent data compression) and modeling (e.g. dynamic traffic assignment and agentbased modeling) that scale in both time and space are now possible. By combining massive amounts of data from realworld sensors and very large road network models, both closed form analytics and emergent behaviors from large scale agent models can be used to build our understanding of urban scale problems. We will address: how these largescale computing resources and big data analytics can contribute to the design of nextgeneration transportation models, analytics, and operational control systems? Workshop Goals:The goal of this workshop is to assemble researchers and practitioners from transportation, largescale computing, and data science communities to address the challenges of integrating nextgeneration technologies into the design of transportation systems. Discussion topics include: ● Big Data Analytics and Data Science with RealWorld Transportation Data: ○ Application of machine learning, specifically LSTM Recurrent Neural Nets, for geospatial data modeling
○ Methods for addressing data veracity and data reduction from both a statistical and learning perspective,
○ Coupling data assimilation and modeling for largescale urban area modeling
● Modeling and Prediction: ○ Modeling the decisions of millions of vehicles and drivers as they interact with service providers, such as navigation apps and installed infrastructure
○ Predicting the impact of new transit options (e.g. modes of mobility) on the system behavior
○ Mitigating the effects of incidents (e.g. natural disasters) on the system
