A Renewable Energy Selection Model for Sustainable Development: Integrating Hesitant Fuzzy Logic and Machine Learning for Renewable Energy Optimization

Achieving SDG 7, universal access to affordable, reliable, and clean energy, is vital for tackling the energy crisis and ensuring sustainable development. In Uttarakhand, despite rich renewable potential, energy access is hindered by challenging terrain, ecological sensitivity, and infrastructure gaps. This research proposes a hybrid decision-making framework combining Machine Learning (ML) with Hesitant Fuzzy Multi-Criteria Decision-Making (MCDM) to identify the best renewable energy options for the region. Five alternatives, solar PV, solar thermal, CSP, mini & small hydropower, and bioenergy, were chosen based on resources and expert input. Using bibliometric analysis in R and the Nominal Group Technique (NGT), criteria were set. The Hesitant Fuzzy AHP assigned weights, while H-FTOPSIS ranked options. Logistic regression enhanced prediction accuracy, and sensitivity analysis tested model stability. Results show solar PV as the most viable choice. The framework supports strategic, evidence-based energy planning for Uttarakhand and offers a scalable, adaptable method for similar regions worldwide.