Tires, bitumen, and road markings are important sources of traffic-derived carbonaceous wear particles and microplastic (MP) pollution. In this study, we further developed a machine-learning algorithm coupled to an automated scanning electron microscopy/energy dispersive X-ray spectroscopy (SEM/EDX) analytical approach to classify and quantify the relative number of the following subclasses contained in environmental road dust: tire wear particles (TWP), bitumen wear particles (BiWP), road markings, reflecting glass beads, metallics, minerals, and biogenic/organics. The method is non-destructive, rapid, repeatable, and enables information about the size, shape, and elemental composition of particles 2-125 mu m. The results showed that the method enabled differentiation between TWP and BiWP for particles > 20 mu m with satisfying results. Furthermore, the relative number concentration of the subclasses was similar in both analyzed size fractions (2-20 mu m and 20-125 mu m), with minerals as the most dominant subclass (2-20 mu m x = 78%, 20-125 mu m x = 74%) followed by tire and bitumen wear particles, TBiWP, (2-20 mu m x = 19%, 20-125 mu m x = 22%). Road marking wear, glass beads, and metal wear contributed to x = 1%, x = 0.1%, and x = 1% in the 2-20-mu m fraction and to x = 0.5%, x = 0.2%, and x = 0.4% in the 20-125-mu m fraction. The present results show that road dust appreciably consists of TWP and BiWP within both the coarse and the fine size fraction. The study delivers quantitative evidence of the importance of tires, bitumen, road marking, and glass beads besides minerals and metals to wear particles and MP pollution in traffic environments based on environmental (real-world) samples