Nashville, TN
Data-Driven Monitoring and Optimizing of Right-of-Way Permits
Organization: Vanderbilt University
Primary Investigator: Ayan Mukhopadhyay
Research Track: Optimizing Resources and Services
NSF Abstract
The objective of this Civic Innovation Challenge (CIVIC) project is to support research on building and piloting a prototype system that uses artificial intelligence and sensor data to automatically detect unpermitted closures. It seeks to enable real-time monitoring of road closures, guide inspectors with optimized routing tools, and support permit staff with better data for future planning. Right-of-way closures—such as the blocking of streets, sidewalks, and bike lanes for construction, delivery, or special events—are an everyday reality in cities. When these closures are not properly permitted or monitored, they disrupt traffic, endanger pedestrians and cyclists, and negatively affect small businesses. While cities like Nashville are working to improve enforcement, they face challenges due to outdated systems, limited inspection staff, and the sheer number of closures. By combining technology with direct input from city officials, field staff and civic organizations, the project seeks to produce a practical, tested system that improves how cities manage public space. Its deployment would support compliance of existing right-of-way closure procedures, recuperation of lost revenue from missing permit application fees, and improve traffic flow and the safety of urban transportation. Importantly, the tools seek to be designed for reuse by other cities and other sensor modalities. The project will release open-source software, deployment guides, and training materials for public access.
The project seeks to advance the automation and optimization of right-of-way closure enforcement through new contributions in machine learning, optimization, and systems engineering. The novelty lies in: 1) the joint integration of anomaly detection and multi-objective inspection routing to support real-time, scalable enforcement; 2) the use of uncertainty-aware learning and human-in-the-loop active validation to improve robustness to noisy, incomplete, or imbalanced urban sensor data; 3) the co-design of a web-based inspector and permit manager interface for operational deployment; and 4) the creation of a generalizable, open-source civic AI framework that can be adapted to infrastructure-constrained transportation departments across the United States.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.