Due to shifting material use in several sectors, such as the automotive sector, the demand for wrought aluminium alloys is significantly increasing. Because of their low weight and desirable mechanical properties, wrought aluminium alloys find their use in many different applications. However, the primary production of aluminium is extremely energy intensive. Therefore, using secondary aluminium yields major environmental benefits. Hence, in order to avoid degradation of the aluminium quality during recycling, sorting aluminium alloys, based on their alloying elements, is necessary. Today, various non-ferrous metal fractions are either still sorted manually in unhealthy working conditions, resulting in either high labour costs, or the export of this waste stream to countries with a lower labour cost. With the emergence of novel spectrometric techniques, such as laser-induced breakdown spectrometry (LIBS) and deep learning computer vision techniques, the technical feasibility of classifying different aluminium alloys has been demonstrated. Therefore, the techno-economic viability of a robotic sorting process, that could be combined with such advanced classification systems, is presented. This study presents the development and evaluation of a robotic sorting system consisting of; a vision system, a conveyor, a SCARA robot and a pneumatic gripper. The vision system recognises the dimensions and positions of the objects on the conveyor and communicates with an innovative sequence planning algorithm. The use of experimental data enables to obtain realistic insights in the sorting efficiencies that can be obtained. The initial economic analysis illustrates the substantial potential of the proposed robotic sorting approach. To overcome saturation of the conveyor belt, two of the proposed systems are assumed to be capable of sorting 20.000 tons of aluminium annually each equipped with 6 robots creating a total added revenue up to 1,95 million euro per year.