AI-Enhanced MCDA-Based Evaluation of Grid Reliability in the Nordic-Baltic Region

This study proposes a novel AI-enhanced Multi-Criteria Decision Analysis framework for monitoring grid reliability in the Nordic-Baltic region. Traditional reliability metrics like SAIDI (System Average Interruption Duration Index) and SAIFI (System Average Interruption Frequency Index) are insufficient for modern power systems due to their inability to capture multidimensional challenges such as renewable energy integration, cybersecurity threats, and climate-induced disruptions. The proposed composite reliability indicator integrates 26 operational metrics into four principal dimensions: grid stress, fault-type trade-offs, cascading failures, and hidden vulnerabilities, using principal component analysis and machine learning-optimized entropy weighting. The TOPSIS (Technique for Order of Preference by Similarity to Ideal Solution) method ranks countries based on their proximity to an ideal resilient grid. Results highlight Sweden, Norway, and Iceland as top performers, while Latvia and Lithuania exhibit higher vulnerability. This AI-augmented framework enhances real-time monitoring, supports adaptive decision-making, and provides a scalable model for global grid reliability assessments.