MIDS Capstone Project Spring 2016

Should I Go

Project Objective

We studied historical data to understand the incremental impact of special events (black swans - these special events act as negative shocks to the system, increasing congestion and reducing traffic velocity) - large sporting events and festivals - on Chicago traffic patterns. Leveraging that information, along with online sources of upcoming events, we improved typical traffic forecasts.

How are we different

Google and Apple maps’ trip duration forecasts are based on typical traffic patterns.  What about atypical days?  They do not incorporate the effect of upcoming special events, like basketball games, conferences, or music festivals.  

Data Sources

Traffic Data

The Chicago Traffic Tracker estimates traffic congestion on Chicago’s arterial streets (non-freeway streets) in real-time by continuously monitoring and analyzing GPS traces received from Chicago Transit Authority buses.  They cover 300 miles of principal arterials and are produced every 10 minutes.  The city provides two historical datasets, the first dividing the city into 29 regions, and the second dividing the city into 1270 road segments.  Congestion estimates provide the observed speed, typically for one-half mile of a street in one direction of traffic.   This data set includes over 10 million records.

Special Event Data

Chicago Park District - Event Permits: The City of Chicago provides open access to its database of event permits.  One of the event types, “Permit - Festival/Performance/12,001+ Event” was particularly interesting for our purposes.  This permit is granted for events on City of Chicago property anticipating more than 12,000 attendees.  Examples include Lollapalooza and the Chicago marathon.  This is exactly the type of extreme event we want to capture with our special event variable. The database spans 2012 through November 2016 - it includes both historical and upcoming events - and includes the reservation start and end date and the park/facility name.  Identifying the latitude and longitude associated with each park/facility would be part of our data processing effort.

Historical Sporting Events: We used basketball-reference.com and pro-football-reference.com  Data sets for the Chicago Bulls and Bears included historical game dates and start times, whether each game was at home or away (we were only interested in home games) and the number of overtimes.

Last updated:

March 30, 2017