Design for Sustainability (D4S) and LifeCycle Assessment (LCA) methods usually focus on one single aspect of sustainability at a time (e.g., environmental issues, ergonomics or costs) and are usually applied when the industrial system is already created, so that only corrective actions can be taken. In this context, the present research highlights the need of predictive methods to design sustainable system, able to provide an early holistic assessment from the early conceptual stages, and defines a set of models of impact able to assess all aspects of sustainability (i.e., environmental, economic and social) by proper key performance indicators (KPIs) from the early design stages. An industrial case study is presented to show the application of the proposed models on industrial manufacturing systems and demonstrate their validity in estimating the global impact on sustainability, including also human factors.