| Sensor | Speed | Flow | Occupancy |
|---|
Operations Report
--
| Corridor | Avg Speed | Vehicles | % Congested |
|---|---|---|---|
| Loading… | |||
| Pollutant | Instant Rate | Today Cumul. | All-Time Cumul. |
|---|---|---|---|
| Loading... | |||
| Metric | Daily | All-Time |
|---|---|---|
| Delay (person-hrs) | -- | -- |
| Delay cost (VOT) | -- | -- |
| Operating cost | -- | -- |
| Total | -- | -- |
| Pollutant | Rate ($/ton) | Daily | All-Time |
|---|---|---|---|
| -- | |||
| Total | -- | -- | |
Before & After Events
Monitor traffic and emissions around planned events before and after they occur. Data is collected silently and stored in a dedicated database.
LATTICE
UNLV Traffic Intelligence Platform
LATTICE is an open-architecture research platform for real-time traffic intelligence and localized emissions quantification, developed at the UNLV Transportation Research Center. It ingests live freeway sensor data from the Nevada Department of Transportation (NDOT) FAST system across the Las Vegas metropolitan area, delivering actionable insights for traffic operations and air quality assessment.
The platform integrates three core capabilities: real-time traffic state monitoring with lane-level analysis, emissions estimation using EMFAC-2021 factors across 20+ pollutants, and deep learning forecasts for multi-horizon speed prediction. Data streams continuously to an interactive dashboard via a live NDOT server connection.
This work is informed by research priorities identified by the Federal Highway Administration (FHWA), the U.S. Environmental Protection Agency (EPA), and the U.S. Department of Transportation (USDOT) for AI-driven traffic management and localized emissions accountability — see Policy & Research Context below.
UNLV Transportation Research Center
Howard R. Hughes College of Engineering
University of Nevada, Las Vegas
T. Zahid, B. Morris, “Using Deep Traffic Prediction for EMFAC Emission Estimation and Visualization,” Proc. 27th IEEE ITSC, Edmonton, 2024.
T. Zahid, B. Morris, “Benchmarking/Limitations of Traffic Prediction with Noisy Field Measurements,” 18th IEEE ICVES, 2024.
Live speed, volume, and occupancy stream directly from NDOT’s permanent freeway sensors every 60 seconds, broken down by individual lanes. This continuous feed provides the high-fidelity, real-time data that federal traffic management programs depend on, eliminating the data-access hurdles that most agencies face.
LATTICE estimates 18 distinct air pollutants at the sensor level every single minute using EMFAC, the EPA-approved model for regulatory analysis. Because transportation is the primary source of U.S. greenhouse gases, this real-time localized measurement fills a critical data gap that regional estimates miss.
An AI model trained on two years of historical NDOT data forecasts traffic conditions 15 to 60 minutes ahead across all major Las Vegas corridors. The model remains completely reliable even when physical sensors report missing or faulty data, overcoming the single biggest barrier to state-level AI adoption.
Every traffic, emissions, and predictive layer is fully open and accessible through a live map with no subscription or licensing fees required. This design prioritizes data equity, putting advanced analytics within reach of smaller agencies and independent researchers rather than just well-funded operations.
Identifies AI-powered predictive traffic management and open digital infrastructure as priorities for transforming U.S. transportation. Notes that 76% of state DOTs cite data quality as the #1 barrier to deploying AI.
Federal program funding advanced transportation technologies that serve as national models for data-driven traffic management.
Transportation is the largest source of U.S. greenhouse gas emissions, at 28–29% of the national total. LATTICE explores localized, real-time quantification as a complement to conventional regional EPA inventories.
Identifies data-driven traffic management, emissions reduction, and equitable access to transportation analytics as central federal objectives under the USDOT. This project's design reflects all three priorities.
Federal research report from the Federal Highway Administration (FHWA) documenting the national need for AI-powered predictive analytics in traffic management — the type of capability LATTICE's speed prediction engine is designed to demonstrate.