1. Home
2. Ideas
3. Traffic Management – The Opportunities a...

Traffic Management – The Opportunities and Challenges for Using Prediction Models for Operations

• Jul 14, 2024

Research Description

Ability to predict traffic condition has been pursued for a long time. However, it is a challenging task to have accurate & reliable prediction models for traffic operations due to many factors, such as data availability and accuracy and model reliability, among other things. These predictions may include traffic congestion, end of queue, hot spots, and incidents. Recent years, probe and CAV data makes it possible to develop, calibrate, and train models for more accurate predictions. Besides traditional prediction models, artificial intelligence (AI) has become more and more useful tool to read in large amount of data, train models, field operating results, and feedback the model to improve the model. This research is focused on the topic of the opportunities and challenges of using prediction models for operations.

Additional Supporting Information

What we hope to learn from this research include: • Goals, needs, and challenges of using prediction models • Guidelines for agencies on how to use predictions models • Data requirements – data availability, quality of data (garbage in garbage out), data reconciliation • Model reliability – reduce false positive to avoid overwhelming agency operation staff • Staff and agency expertise • Testing & validation of prediction models

Benefits: • Reduce congestion & Emissions • Reduce operating cost • Reduce traffic breakdown & serve more vehicles • Save lives – reduce incidents • Improve agency efficiencies • Help agencies understand so that they are more willing to use prediction models for their daily operations • Attract more talents from fields such as civil engineering, computer engineering, and other areas to improve prediction models

---

Submitted By

Jim Katsafanas

Michael Baker International

412-269-4635

Attachments