Surge Prediction

Why We Made This

Problem

• Using ride-sharing apps can be unpredictably expensive.
• Local environmental factors (Weather | Social & Sporting Events) cause demand for drivers, and fares increase to meet this demand.

Solution

• Local events and weather predictions are compared in a logical way, to provide users with the likelyhood (%) of a fare surge for the next three hours.
• Users are able to call a Lyft directly from the site.

How Does It Work?

Information Gathering

• First, data is gathered from the following API
  • Weather Underground (Weather info: Percepitation, Incliment weather)
  • Google Maps && Places (Traffic Data)
  • Ticket Master (Event Start & End Times | Number of Events | 5 Mile Radius)
  • Lyft

Logic

• Logic for % Chance of Fare Surge:
  • Current weather, and current + 2 hours
  • Local events beginning and ending within the hour, and the following two hours
  • Predicted likelyhood offare surge increases as incliment weather increases, or with presence of events. Both conditions occuring at the same time dramatically increase the likelyhood of a surge in pricing.

Tech Used

• JavaScript / jQuery
• Firebase (Data Storage)
• Chart.js (Graphical Display of Prediciton)
• Ajax