Should I Go?


Aggregating special event data to avoid black swans in traffic


Problem

Our Solution



We propose a service that enhances traffic predictions by aggregating special event information and incorporating “black swan” events into forecasts.

We studied historical data to understand the incremental impact of special events - large sporting events and festivals - on Chicago traffic patterns. Leveraging that knowledge, along with online sources of upcoming events, we can improve traffic forecasts by adding a new variable to the model: special events.

This additional dimension of information allows Chicago residents and visitors to make better informed choices when planning events. Not just “What is the typical commute time for that neighborhood?” but “Given upcoming sport and entertainment events, what will traffic congestion look like next Wednesday?”

Data Science

Demo

Technology

Future



Expand across multiple cities and include more events

Event Notification to subscribers on the common route

Explore RNN Models for improving the model

Better methods for outlier detection

3 tier architecture