<p>In sustainable manufacturing, the trade-off between economic efficiency, environment, and human well-being plays a very important role. Machining is a one of the most important processes for the realization of a product, but it is heavily loaded with high-energy consumption and noise. Few studies have addressed both objectives simultaneously while accounting for surface requirements. In this study a predictive modelling and multi-objective optimization approach was developed for Computer Numerical Control (CNC) milling of 2017&#xa0;A aluminum alloy, while minimizing the energy consumption and noise emission under finishing-oriented conditions. Experimental tests were carried out under contouring and surfacing operations with variations of ‘cutting speed’, ‘feed rate’, and ‘depth of cut’. Later, a multi-objective optimization problem was solved with the use of Non-dominated Sorting Genetic Algorithm II (NSGA-II) and then decision-making was performed by Technique for Order Preference by Similarity to Ideal Solution (TOPSIS). Surface quality was considered through a predictive constraint to ensure finishing -level conditions. Contouring Pareto front showed that ‘high cutting speed’ along with low ‘feed rate’ and high ‘depth of cut’ was able to minimize energy consumption and noise emission effectively without compromising on surface quality constraints. On the other hand, in surfacing operations, there was an obvious need for a trade-off: while a low feed rate consistently improves the surface roughness, a relatively lower value of cutting speed was selected in order not to deviate noise and energy beyond the limits. Similarly, this approach reflects the sustainable manufacturing concept because the optimized parameters also contribute indirectly to lower operational costs and associated <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\({\text{C}\text{O}}_{2}\)</EquationSource> </InlineEquation> emissions.</p>