Open Science Data Challenge

The world is asking for change

Sustainability is the most pressing issue of our time for the environment, employees and consumers, society and providers of capital.

At EY, we believe that value-led sustainability is bigger than any one business. It’s everybody’s business. 

We also have the convening power to bring everyone along for the journey: where everyone has a role to play in realizing new solutions that work beyond business to society as a whole.

Held annually, the EY Open Science Data Challenge produces new IP that drives more intelligent, information-driven and better decisions in mitigating global sustainability issues, such as climate change. EY makes this IP available to NGOs, governments and scientists for non-commercial purposes.

Working together, we can scale our impact through data, AI and technology to collectively build a sustainable future for society and the planet.

The 2023 Challenge

At EY, we want to enable the creation of open-source solutions that could help save 800 million people from hunger. Now closed, the 2023 Open Data Challenge generated over 13,000 registrations from students at top universities and early-career professionals from organizations across the globe.

With the mission of harvesting data to help solve world hunger, participants used satellite data from Microsoft’s Planetary Computer to predict the effects of climate change on rice production, helping farmers, communities and countries grow more food. Through collaboration with leading professionals, challenge participants honed their skills in programming, algorithmic processes and artificial intelligence to create new tools to help feed the world. Their efforts will contribute to the development of a better, more predictable and more sustainable future. 

A distinguished panel of judges is currently evaluating submissions and expects to announce winners in the coming weeks.

A look back at the 2022 Data Challenge

Getting started

This year’s challenge has two components:

Challenge 1: Build a system to detect fire-edges in infrared linescan images

Challenge 2: Build a system to map the location and behaviour of fire-edges in satellite images

Both challenges are designed to help streamline the process of bushfire mapping for bushfire authorities. Using data science and advanced analytic tools, your task is to map the location of the fires based on observations in the provided data. Once you have submitted your results, you can continue to refine your model over time to achieve a higher score.

Who can enter

Any university student with an interest in data science can enter and participate. If you have less than two years of professional experience then you can also join the Challenge, but you will need to pay to use Azure.

How to enter

The platform to submit your results will open on 24 March 2021.

You can enter the challenge as an individual or as a team (maximum four people).

You can make multiple submissions (maximum one per day) with your account before the challenge closes.

Submission requirements and evaluation

Note that participants are invited to use whatever technologies and/or methodologies they feel would be most appropriate for the challenge at hand. Your results must be submitted on the platform.

If you are selected as a finalist, you will be provided instructions on how to submit your results and supporting documentation, which should include an explanation of methodologies and algorithms used, along with any code leveraged, assumptions made and any insights identified.

Finalists will also be required to submit a video presentation of their work. Evaluation criteria will be shared once the finalists are announced.

You will also be required to provide any code and external datasets leveraged, so be sure you have used only publicly available data and that your code and methodology is robust and repeatable.

Deadline

All submissions must be made by midnight, 15 June 2021 (GMT). The top 60 global finalists will be notified and required to present a video with their findings, methodologies and assumptions, and provide any artefacts such as code base to the judges as requested. Finalists will also be required to attend the global virtual awards ceremony in the first half of July 2021.

Register here