Harnessing data science for real-world solutions

As part of the ongoing C-DICE Grand Challenge, C-DICE partnered with the Alan Turing Institute to harness the power of data science in a Turing Data Study Group (DSG). PhD students and postdocs from around the world recently took part in a week-long online DSG, a collaborative “hackathon” in which participants use data science to find novel solutions to real-world problems.

The C-DICE Challenge

The UK Government’s net zero strategy requires phasing out non-renewable power plants while expanding renewable energy sources such as wind, solar, and nuclear. However, determining when and how to decommission, repurpose, or enhance existing energy infrastructure remains a critical challenge. The goal of this challenge was to design a strategic roadmap that balances energy demand, cost efficiency, and workforce capacity over the next 25 years.

The approach

To understand the energy landscape in 2025, participants used the publicly available JRC Power Plant Database that includes details of power plants such as locations, capacities, and fuel types. To learn the milestones that should be met along the way, we used ESME (Energy System Modelling Environment), that provides two long-term energy transition scenarios: clockwork and patchwork. The decisions that could be made to modify the energy landscape are:

• Decommissioning fossil fuel plants at optimal times.
• Repurposing certain facilities into renewable assets.
• Enhancing existing renewable capacity.
• Expanding new renewable energy infrastructure

Results

The participants used Mixed Integer Linear Programming (MILP) to develop an optimisation model that determines the best sequence of actions to meet the milestones in the scenarios. The model generated by the participants can return the optimal sequence of actions for any scenario. The model successfully optimised energy transition pathways across different scenarios. Some of the findings include:

• Rapid transitions require realistic workforce planning: Without constraints, the model aggressively decommissions fossil fuels early, which may not be feasible due to infrastructure and workforce limitations.
• Gradual phasing balances feasibility and cost: By limiting the number of projects per time period, we observed smoother transitions aligning with real-world constraints.
• Wind and nuclear energy play a major role: Both energy sources emerged as dominant alternatives for replacing fossil fuels in the optimisation framework.

Conclusion and future work

The framework developed provides a data-driven strategy for guiding policymakers and energy planners in their decision-making. Future improvements could include incorporating uncertainties, such as policy shifts, energy price fluctuations or energy demand in the future. The combination of optimisation techniques with real-world energy data can help us meet the desired scenarios in the most cost-efficient way. C-DICE will further develop the outputs from the DSG when the final report is published.  

By Tereso del Río Almajano and Sharon Henson

About the Alan Turing Institute

The Alan Turing Institute, headquartered in the British Library, London, was created as the national institute for data science in 2015. In 2017, as a result of a government recommendation, it added artificial intelligence to its remit.  The Institute is named in honour of Alan Turing (23 June 1912 – 7 June 1954), whose pioneering work in theoretical and applied mathematics, engineering and computing are considered to be the key disciplines comprising the fields of data science and artificial intelligence. 

In 2023 the Institute launched an open university network, providing all UK universities with an interest in data science and AI the opportunity to engage and collaborate both with the Institute and its broader networks. 

Getting involved in the Institute’s Data Study Groups (DSGs) offers the chance to tackle real-world data challenges, collaborate with top researchers, and potentially kick-start larger projects. It’s a great way to apply your skills, innovate, and network within the data science community.