System for analyzing user driving behavior

  • project: octoU
  • client: Octo Telematics
  • year: 2015
  • website: octou.com

Mobile Application & Analytics Platform

challenge:

Our partner – a large Italian telematics company – sought to build a powerful mobile application to analyze driving behavior of users.

Our challenge was to implement the criteria and put them into a seamless, user-friendly application.

With the app already in hand, we focused on engineering a platform for analyzing, querying and visualizing a set of data provided by the system.

tools used:

Java 8
Spring
Cassandra
ActiveMQ
MySQL
Spark
Scala
Angular JS
We developed a feature-rich app that allows users to measure their driving skills.

solution:

The system is based on heterogeneous data, such as real-time GPS data, weather and road conditions, traffic congestion, speed limits or drivers’ lifestyle information. Each users’ trip is assigned a score that describes how safely the user was driving. The algorithm detects hazardous events, such as speed limit violations, harsh acceleration, braking, and cornering.

The application contains social features, such as commenting and liking friends’ trips, collecting badges, receiving tips etc.

Gamification
Driving tips
Social features
Data Storing
Driving Monitoring

The system is distributed and is deployed on multiple nodes that asynchronously process the incoming data. The platform can be used to discover correlations between measured parameters. It allows to monitor the driving style of a particular driver. Data processing is done using Apache Spark with Cassandra as a non-relational data storage.

impact:

OctoU is a silver winner of Best Mobile App Awards 2015

OctoU won the FWA as Mobile App of the Day.

OctoU App achieved:

  • 130.000 downloads
  • 70.433 active users
  • 237 countries with registered users
  • 9.300 webpage views
  • 2.319 webpage active users
  • 14.025.982 kilometers driven
  • 937.157.648 GPS points collected
  • 17.357.520 minutes behind the wheel