Hackathon #HackZurich: what useful solution can you code for drivers in 40 hours?

A recap of 3 days of hacking in the Technopark in Zurich. More than 500 coders joined #HackZurich, one of the largest and most prestigious hackathons in Europe. Efficient working weekend of the year!

Last weekend we — Bright Box team — supported our first big hackathon by providing the anonymous driving history of 1000 millennials from 6 countries for participators to retrieve and share valuable insights and to turn them into products or services for the Challenge organized by Zurich Insurance and Esri. The challenge from Zurich Insurance & Esri was named «Millennials on the Move». The idea of the challenge was in by the anonymous data from Bright Box should helping to find out the driving behavior of the millennials and how it can help improve and protect the lives of millennials on the move. Coders also got access to ArcGIS; a complete mapping and analytics platform for developers: use location intelligence to the advantage and combine demographic and lifestyle data with millennials’ movements. And I want to say that we had a blast!

The workshops

Day 1 was fully packed with workshops from 20 big companies. They addressed the subjects like safety, create new tech-driven insurance solutions, sustainability, battling fake news, GDPR.

The attractiveness of our workshop was that participants could work with real anonymized data. We are focused on empowering the business of our customers through the value of connected vehicle data. And we hoped to find among the participants of the hackathon who will be interested to make automotive data as valuable as possible to drivers. And we did. 80 hackers were with us to the workshop «Millennials on the Move».

Research shows that millennials are horrible drivers. Texting, emailing, chatting behind the wheel, running red lights, eating and drinking are just a few of the reckless acts millennials do while driving. The first challenge was to retrieve and share valuable insights about how millennials really drive and how new solution can help improve and protect the lives of millennials on the move. And the second aim was to turn the insights into products or services.

During our workshop, we showed what specific date sets participants will work with. One year’s data on millennial drivers compared with the all-age group at the same time, in the same city. And we provided access to ArcGIS to help rediscover all the ‘behind-story’ data given any GPS location and time stamp, such as demographic and lifestyle data. Participators of the workshop asked about the availability of accelerometer data, what is the frequency of the data (interval between points and time). As a result, our challenge was attractive for 8 teams. And it’s definitely a success.

Developers are rock!

It was so exciting to see all those cool projects built in a little over 40 hours! By the end of Hackathon, 8 teams had a working proof of concept and presented their hacks at challenge «Millennials on the Move». Some of them were very prepared and already had the demo version and published the code on GitHub.

Below you can find the overview of some most interesting of alls by our opinion.

CAR TaiLOR

Car tailor uses advanced analytics to aggregate behavior as well as personality data from various sources. These aggregated user profiles allow us to recommend suitable products, that match the lifestyle and personality of customers.

SafetyMatters

The app has two parts. One alerts the authorities of dangerous junctions where many people have sharp changes in speed and have many car accidents in that area. The second part uses a personalized learning algorithm that learns where the driver needs to be alerted before places he is driving in an unsafe manner. This gives a comprehensive solution to risk mitigation of car accidents.

Safety Matters Demo

Safetify

A prototype app uses a machine learning model that takes traffic, driver, road and weather data into account, predicts how dangerous a particular road segment is and alerts users. The probability of a segment to be a danger zone was computed with neuronal networks. The area of the danger zone was computed by using Gaussian Process Regressot to smooth out the probabilities and provide a contour.

Savfe

The app detects whether the vehicle owner is driving and looking at the smartphone. If he in the app his screen will start flashing warning colors (yellow and red) depending on how dangerous the situation currently is. If he is distracted by another app, the app shows a message so that he should take care. If the vehicle owner didn’t use his device while driving, that’s great! And the app will reward him for it with a special currency. And he gets even more of it if he manages to ‘uphold your streak’ and doesn’t use his smartphone for consecutive trips. And he can spend his points to get back a part of his monthly insurance cost.

And our workshop “Millennials on the Move” made by Esri, Zurich, and Bright Box, was won by four ladies with the project “Safety Matters”. Girls power!

The hackathon was the first for the Bright Box. And it was amazing. This event for us wouldn’t have been possible without the invitation of Zurich Insurance Group. And we are grateful for the opportunity to test such creative ideas on our data that once again proves their value.

--

--

Alexander Dimchenko
Bright Box — Driving to the future

Chief Strategy Officer at Bright Box, global vendor of Connected Vehicle Platform - Remoto (www.remoto.com) https://goo.gl/K1E8NQ Download our free white paper!