CAT Presentation | Muhammad Tanveer Ul Islam
Student: Muhammad Tanveer Ul Islam
Title: Towards Sustainable Local Energy Initiatives: Integrating Operational Research and Machine Learning
Date: 02/09/2025
Time: 15h00
CAT members:
Chairperson: Professor Inês Isabel Carrilho Nunes (IST)
Supervisor: Professor Miguel Brau Canadas Alves Pereira (IST)
External Member: Professor Ana Sara Rodrigues da Costa Domingues (UÉvora)
External Member: Professor Carla Margarida Saraiva de Oliveira Henriques (CBS | ISCAC)
Brief description of the research work: The transformation of renewable energy systems is a key objective for sustainable environmental and economic development, particularly at the municipal and community levels. Achieving this transition requires innovative tools to address decentralized renewable energy adoption's technical, social, and economic challenges. This research aims to explore these challenges within local energy planning, demonstrating how decision-making methodologies and machine learning (ML) models can accelerate the shift to renewable energy while ensuring efficiency, feasibility, and stakeholder engagement. The study is structured into three interconnected phases, each contributing to the development of practical methodologies for energy transitions. Phase 1 proposes a Multi-Criteria Decision Analysis (MCDA) framework for ranking municipal Photovoltaic (PV) projects by systematically evaluating their technical, economic, environmental, and social feasibility. This framework is designed to improve transparency in deStainableking by incorporating stakeholder perspectives, balancing multiple conflicting criteria, and ensuring optimal project selection. The fraenhancing decision-making by providing more precise, data-driven insights that improvealidate its applicability and effectiveness. Phase 2 focuses on developing ML models to predict energy demand and electricity production, particularly within Renewable Energy communities (RECs). These models will utilize historical energy consumption and generation data, applying advanced algorithms to improve forecasting accuracy, optimize energy exchanges, and enhance resource utilization within RECs. The ML models will be trained and validated using real-world energy datasets to ensure reliability in practical applications. Phase 3 integrates the MCDA framework with ML-based predictive models to support Sustainable Energy Planning (SEP) and decentralized renewable systems. This integrated approach combines qualitative and quantitative assessments, improving decision-making by offering more precise, data-driven insights that enhance predictability and adaptability in energy management strategies. The combined framework will be applied to real-world case studies to assess its effectiveness in guiding municipal and community-level energy transitions. The proposed methodologies are expected to empower local governments, policymakers, and communities to make informed, scientifically backed decisions that address technical, economic, and societal challenges in the transition to renewable energy. By bridging the gap between theoretical models and practical implementation, this research contributes to advancing local energy planning and offers concrete recommendations for designing responsive, community-driven energy solutions. Ultimately, this study presents methodological advancements at the intersection of decision analysis and predictive modeling, contributing to the broader research effort to develop efficient, scalable, and adaptable frameworks for sustainable energy transitions.
Research Unit: Centre for Management Studies of IST (CEGIST)