<p>Project portfolio selection and scheduling are critical decision-making processes in project-oriented organizations, directly impacting resource allocation, financial performance, and long-term organizational success. In recent years, the emphasis on sustainability has intensified, compelling organizations to balance economic performance with environmental responsibility. This study addresses the need for environmentally conscious project portfolio selection and scheduling by proposing a novel optimization approach. A multi-objective mixed-integer linear programming (MILP) model is developed to simultaneously maximize portfolio profit, minimize energy consumption, and reduce environmental pollution. This comprehensive framework enables organizations to optimize project portfolios while aligning with sustainability goals and regulatory requirements. To address the challenge of solving large-scale instances, a new hybrid solution method is proposed, integrating an enhanced Strength Pareto Evolutionary Algorithm with deep learning techniques. This approach boosts computational efficiency, elevates solution quality, and equips decision-makers with more reliable tools to navigate competing objectives. Computational experiments demonstrate the effectiveness of the proposed approach, showing significant improvements in environmental impact reduction without sacrificing financial performance. The results validate the model’s practical applicability in real-world scenarios, particularly for organizations adopting green management practices. Ultimately, this study contributes to green project management by offering an innovative decision-making framework that supports sustainable development goals while helping organizations achieve long-term business success with minimal environmental impact.</p>