Big Data

Big data analytics assembles, stores, and analyzes large amounts of relevant data to monitor performance, and gain insights regarding interrelationships of various components. At the UFTI we are using various methods and tools for transportation data visualization, video analytics and crash risk, Connected and Autonomous Vehicle (CAV) operations, traffic signal control optimization, and transportation planning procedures, among others.

Selected projects, related UFTI events and resources, and bibliographical information of published papers are provided below.

Example Projects

Title Principal Investigator/Co-PIs Agency/Source Description
Data Management and Analytics for UF Smart Testbed Dr. Sanjay Ranka, Dr. Anand Rangarajan, Dhruv Mahajan, Tania Banerjee FDOT The University of Florida (UF) and its Transportation Institute (UFTI), the Florida Department of Transportation (FDOT) and the City of Gainesville (CoG) are coordinating the development of a smart testbed on the UF campus and adjoining city streets. In this project we analyze the key requirements for developing a IoT Data Warehousing Platform that seamlessly  collects, stores and provides access to the massive amount of data generated by the different types of sensors such as loop detectors, cameras, radar, LIDAR, etc. The platform will also support deep analytics to discover insight and for decision making.
Optimizing Traffic Grids with Machine Learning Dr. Sanjay Ranka, Dr. Anand Rangarajan, Rahul Sengupta, Hoda Shajari FDOT In this project, we investigate the use of Machine Learning(especially Reinforcement Learning) to the problem of optimizing traffic flow in dense urban grids with signaling mechanisms. We use microscopic traffic simulators(such as SUMO, VISSIM etc.) and recorded field data(Loop detector data, vehicle probe data etc.) to build realistic models of typical peak hour traffic scenarios in hypothetical and real-world grids. We then employ advanced Machine Learning techniques(incl. Multi-agent Reinforcement Learning methods) to optimizing signaling behaviors and achieve optimal traffic flow.
Large-scale Traffic Corridor Identification using Big Data Dr. Sanjay Ranka, Dr. Anand Rangarajan, Rahul Sengupta, Vineeth Kamisetty FDOT In this project, we investigate and deploy algorithms to detect recurrent traffic flows in urban traffic grids based on large-scale pre-recoded historical traffic data. We warehouse and preprocess large datasets and then use spatiotemporal clustering techniques on them to detect underlying corridors of traffic flows.

More projects can be found on the TRID database at


Lily Elefteriadou, Ph.D.

Professor Civil and Coastal Engineering

Sanjay Ranka, Ph.D.

Professor, Fellow IEEE, Fellow AAS Computer & Information Science & Engineering

Siva Srinivasan, Ph.D.

Associate Professor Civil & Coastal Engineering