Page 12 - Annual Report - interactive demo

Research
Accomplishments and findings from completed and ongoing projects
An AI Platform for Network-wide Congestion Detection and Prediction Using Multi-source Data
Principal Investigator: Yinhai Wang, UW
The advancement of smart traffic sensing, mobile communication, and artificial intelligence technolo-
gies has stimulated massive growth in the amount of transportation data available. Simultaneously, increased computation power enabled by advanced hardware and the rise of artificial intelligence (AI) technologies, espe- cially in the deep learning field, has made it possible to comprehensively utilize transportation big data. In this project, the research team built a prototype transporta- tion AI platform for solving challenging transportation problems, which require large-volume high-dimensional transportation data and complex models. This AI plat- form is capable of providing standardized datasets and novel deep learning-based models for specific problems, and its novel architecture enhances the efficiency of data processing, management, and communication and increases the platform’s computational power. It is ca- pable of evaluating the traffic prediction performance of various implemented models by comparing and visual- izing the prediction results tested on multiple real-world, network-wide traffic state data sets.
Monitoring and Control of Overweight Trucks for Smart Mobility and Safety of Freight Operations
Principal Investigator: Hani Nassif, Rutgers
In New York City, major decisions must be made to al- locate limited funding for repair, maintenance, and reha- bilitation of the infrastructure network. These decisions should be based on integrating various sources of infor- mation to control the infrastructure systems and their deterioration using structural models coupled to traffic modeling at the network level, to help perform econom- ic forecasting and life-cycle cost analysis. This project aims to monitor the impact of overweight trucks on the bridges and pavement infrastructure under the jurisdic- tion of NYC Department of Transportation (NYCDOT). Using weigh-in-motion data from NYC and New Jersey, the research team quantified the preliminary economic impact of overweight vehicles on bridges as dollar per overweight-ton per deck area per trip for three case studies in NYC and found that the unit damage cost is constantly higher for NYC than for most NJ highways. The results from a preliminary analysis on selected cor- ridors in NYC estimate the impact on pavements in the range of $0.0345 and $0.0698 per equivalent single axle load lane-mile on an interstate highway and between $0.117 and $0.648 per ESAL-lane-mile for local roads.
AI platform user interface (top) and Seattle-area Flow of overweight trucks near a New York City Bridge. roadway networks from which the platform’s
datasets are drawn (bottom).
11 C2SMART Center Annual Report