University of Twente Student Theses
As of Friday, 8 August 2025, the current Student Theses repository is no longer available for thesis uploads. A new Student Theses repository will be available starting Friday, 15 August 2025.
Predicting customer churn in a liberalized energy market : A Machine Learning and Explainable AI approach
Gülbey, Baran (2025) Predicting customer churn in a liberalized energy market : A Machine Learning and Explainable AI approach.
|
PDF
3MB |
| Abstract: | In liberalized energy markets such as the Netherlands, where all consumers receive energy through a centralized infrastructure, suppliers face high competition not on product quality but on pricing, service, and customer experience. This dynamic leads to high customer churn rates, causing significant financial challenges for energy providers. This thesis investigates how Machine Learning (ML) and Explainable AI (XAI) can be used to proactively reduce customer churn within such markets. Using the Design Science Research Methodology (DSRM), a structured process was developed and validated through a real-world case study involving customer data from a Dutch energy supplier. The resulting artifact includes a churn prediction model, with XGBoost performing most effectively, combined with XAI techniques such as SHAP and LIME to improve interpretability. These predictive insights were incorporated into a six-step Proactive Churn Retention Process (PCRP), which enables energy suppliers to integrate customer churn prediction into their internal processes and customer retention strategies. Expert validations confirmed the practical relevance and feasibility of the proposed approach. This thesis contributes both a predictive model and a business-oriented framework that bridges the gap between technical insight and strategic application. Customer churn prediction serves as a valuable step toward better understanding customer behavior, allowing companies to deliver a better customer experience. While the research is limited by data scope and evaluation scale, it offers a solid foundation for future work in data-driven customer retention. In markets where switching is easy and competition is based on pricing, understanding and anticipating customer behavior becomes a key differentiator. |
| Item Type: | Essay (Master) |
| Faculty: | EEMCS: Electrical Engineering, Mathematics and Computer Science |
| Subject: | 54 computer science, 85 business administration, organizational science |
| Programme: | Business Information Technology MSc (60025) |
| Link to this item: | https://purl.utwente.nl/essays/107818 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page