*3.1. Composite Materials*

ML and AI algorithms are already widely used in the field of composite materials for tribological applications. Generally, there has been a remarkable growth in the large-scale use of materials made from two or more constituent materials with different physical or chemical properties, for example a fiber and/or filler reinforced polymer (PMC), ceramic (CMC) or metal (MMC) matrix composites. The advantages of these materials lie especially in the high strength-to-weight as well as stiffness-to-weight ratios [30]. For a general overview of tribological properties of different composites in dependency of contact and/or environmental conditions, the interested reader is referred to various review articles [31–33]. A major field which already exploited ML approaches to a greater extent have been wear-resistant composites with polymer matrix, for example thermosets such as epoxy or polyester [34] as well as thermoplastics [35], e.g., polyamide (PA), polyphenylene sulfide (PPS), polytetrafluoroethylene (PTFE), polyethylene (PE), polyether ether ketone (PEEK) [36,37], or polypropylene (PP) [38].
