RESEARCH & DEVELOPMENT ENGINEER
Dr. Patire is a Program Leader and Research and Development Engineer for California Partners for Advanced Transportation Technology (PATH) at the University of California, Berkeley. He earned his master’s and bachelor’s degrees in Electrical Engineering and Computer Science from the Massachusetts Institute of Technology and completed his Ph.D. in Civil and Environmental Engineering at UC Berkeley, focusing on traffic flow theory and bottlenecks.
Dr. Patire has a diverse range of experiences at the intersections of traffic engineering, big data, software and hardware development, data analytics, and machine learning. He has published his work in academic journals, contributing to knowledge in areas such as real-time data fusion for traffic state estimation, traffic flow theory, and bottleneck-activation mechanisms. In addition to his experience building and calibrating large-scale traffic microsimulation models (1000+ lane-miles), Dr. Patire has years of experience with public sector agencies on data quality, data cleaning, and processes to extract actionable information to support traffic management and business decision making.
His recent projects include:
Erroneous HOV Degradation. Applied a range of machine learning methods, including both supervised classification methods, and unsupervised anomaly detection methods to detect HOV misconfiguration errors in the Caltrans data pipeline. Demonstrated that the reported degradation of HOV-lane facilities is greater than the actual degradation, and that potential exists to use machine learning to improve the performance measures that inform operating policies for HOV-lane facilities.
Multiple ICM Corridor Management. Formulated strategies in which multiple, adjacent Integrated Corridor Management (ICM) projects may work together. Identified situations in which traffic management decisions on one corridor may affect a nearby corridor.
Hybrid Data Implementation. Created a strategic roadmap for Caltrans to integrate third-party, travel-time data with Performance Measurement System (PeMS) data to reduce costs and increase coverage of traffic monitoring, improve existing deployment of point-based sensors, and provide a methodology to calculate delay from a flexible mix of data types. Investigated methods using traditional traffic theory, adaptive smoothing, and machine learning. Demonstrated the performance of the algorithms for a range of operating conditions and infrastructure categories.Dr. Patire also brings two years of experience in satellite telecommunication systems engineering and three years of experience in embedded systems hardware design to the PATH team. For the past several years he has helped to bring a number of PATH projects to completion, including Mobile Century and Mobile Millennium. Along the way, he has been a team leader on large scale problems that involve GPS-based probe data, data assimilation, and flow models in the context of integrated corridor management. In his previous position, Dr. Patire was responsible for the analysis, modeling, and simulation of the I-210 corridor as a part of PATH’s Connected Corridors Program.