26/11/2025
💡 Fresh Insights from MAKE
Welcome to read the recent paper "A Four-Dimensional Analysis of Explainable AI in Energy Forecasting: A Domain-Specific Systematic Review" by Vahid Arabzadeh and Raphael Frank.
This systematic review examines 50 peer-reviewed studies (2020–2025) that apply Explainable AI to load, price, and renewable-generation forecasting. Using a PRISMA-guided process, the work introduces a dual-axis taxonomy and a four-dimensional evaluation framework covering global transparency, local fidelity, user relevance, and operational viability. The analysis reveals that XAI usage in energy forecasting is highly domain-specific rather than uniform. Three distinct paradigms emerge: a user-centric approach in load forecasting, where explanations support operators and consumers; a risk-centric paradigm in price forecasting, where XAI is used to validate economic logic and manage financial volatility; and a physics-centric paradigm in renewable generation forecasting, where explanations must reflect meteorological and physical principles.
The review concludes with actionable guidelines and identifies future research priorities, including standardized robustness benchmarks, improved user-cantered design, and greater emphasis on operational feasibility for real-world deployment.
🔗 Read it here: https://www.mdpi.com/2504-4990/7/4/153
MDPI University of Luxembourg
Despite the growing use of Explainable Artificial Intelligence (XAI) in energy time-series forecasting, a systematic evaluation of explanation quality remains limited. This systematic review analyzes 50 peer-reviewed studies (2020–2025) applying XAI to load, price, or renewable generation forecast...
