Prediction, Expectation, and Surprise: Methods, Designs, and Study of a Deployed Traffic Forecasting Service

TL;DR

This paper discusses the development and deployment of JamBayes, a traffic forecasting service for Seattle, focusing on model accuracy and the ability to predict unexpected traffic surprises.

Contribution

It introduces a deployed traffic forecasting system that combines predictive accuracy with surprise detection, advancing practical traffic management tools.

Findings

JamBayes is actively used by over 2,500 users.

Models effectively forecast traffic flow and congestion.

Research on identifying and predicting traffic surprises.

Abstract

We present research on developing models that forecast traffic flow and congestion in the Greater Seattle area. The research has led to the deployment of a service named JamBayes, that is being actively used by over 2,500 users via smartphones and desktop versions of the system. We review the modeling effort and describe experiments probing the predictive accuracy of the models. Finally, we present research on building models that can identify current and future surprises, via efforts on modeling and forecasting unexpected situations.