Data-driven Decision Support Applications for Energy Transition

As this transition progresses, decision-makers at all levels face increasingly com-plex choices that require well-informed strategies. Traditional decision-making approaches, while valuable, often struggle to adequately address the multifaceted nature of energy transition challenges. To navigate these challenges effectively, there is a growing need for data-driven decision support technologies that can enhance and complement conventional methods. This shift towards data-driven approaches is facilitated by the increasing availability of vast amounts of energy-related data from smart grids, internet of things devices, and other digital technologies. This wealth of data provides new opportunities for enhanced decision-making support through improved modelling, forecasting, and optimization capabilities.

This PhD thesis investigates the role of data-driven decision support tools in facilitating the energy transition. While the research focuses on three representative decision situations – selecting heating technologies for buildings, forecasting energy usage and generation, and establishing energy sharing communities – its in-sights are applicable to a broader range of energy transition challenges.

The aim of this thesis is to investigate how data-driven decision support tools are designed and applied in energy transition contexts, examining the interconnections between tool characteristics, decision contexts, and stakeholder needs. Two primary research questions guide this investigation: Firstly, what are the critical factors influencing the design and functionality of data-driven decision support tools for selected energy transition contexts? Secondly, how can data-driven decision support tools be effectively applied in different energy transition decisions?

To address these questions, the thesis provides a review of decision support approaches with practical applications, including multi-criteria decision-making, machine learning models, and optimization techniques. It evaluates their effectiveness in various decision-making situations, considering both the quality of output and the usability of the tools. This dual focus acknowledges that for a tool to have a real impact on energy transition, it should not only provide accurate results but also be accessible and practical for users.

The findings contribute to the field by identifying key factors in tool design, pro-posing guidelines for application, and offering insights into the integration of multiple decision support techniques. By investigating data-driven decision support in representative energy transition situations, the thesis provides a foundation for in-formed decision-making in a wide range of energy transition contexts. The critical role of adapted, transparent, and user-friendly data-driven decision support tools in navigating the complexities of those specific decision problems is underscored.