The behaviour of material flows and waste treatment machines, as well as their mutual influence in plants for the treatment of mixed solid waste, is very little known in the waste sector. Research into these complex interrelationships can provide the basis for white-box models and help advance digitisation in waste treatment. The present doctoral thesis is based on extensive test series which carried out in a specially constructed semi large scale pilot plant (technical line 4.0) and in several large-scale waste treatment plants. In total, more than 1,500 t of municipal and commercial waste were used as test material in more than 60 individual tests. The interactions between machines and between machines and processed material were examined. The focus was on the basic process steps for waste treatment like shredding, sieving, ballistic separation and sensor-based sorting. Among other things, the effects of modification of various shredder parameters, such as cutting gap and shaft speed, on throughput, grain distribution and energy consumption during the coarse shredding of mixed commercial waste with different shredder cutting units were determined. The influence of a screening step downstream of the pre-comminution on the achievable quality of rejected recyclable materials and solid recovered fuels was examined with the aid of industrial screening machines (drum screen and circular vibrating screen). It could be shown how pollutants and fuel parameters are distributed among the individual grain size classes when the sieve perforation changes. When generating a 3D plastic pre-concentrate from mixed commercial and municipal solid waste, it was possible to demonstrate that pre-screening using a drum screen positively affects the downstream processes of ballistic separation and sensor-based sorting. Real-time material flow monitoring is of central importance in the digitisation of waste treatment plants and the development of the future waste treatment plant. In all test series (except for tests with a circular vibrating screen), special attention was therefore paid to material flow monitoring, especially volume flow and mass flow measurement, and the most representative possible sampling of the inhomogeneous material flows used. It could be shown that material flow fluctuations can harm the performance (purity, output, incorrect output) of sensor-based sorting machines.