This thesis studies ESG-aware portfolio choice as a problem with three goals: increase expected return, lower volatility, and cut ESG risk. The setup stays close to practice: long-only portfolios, full investment, limits on single positions, and a fixed range for the number of holdings. The tests use FTSE MIB stocks. Historical prices provide return and risk estimates, and ESG risk scores represent non-financial exposure. Tabu Search drives the optimization, it tries many portfolios, uses short-term memory to avoid repeating the same moves, and keeps only the best trade-off portfolios. The non-dominated solutions are shown on the Pareto front. The analysis also comments on diversification and concentration, and notes that changes in index membership can shift the investable set over time.