Accurately predicting friction in sliding interfaces is critical for many engineering applications. It is known that the coefficient of friction evolves with sliding conditions, material interactions, and contact history. Rate-and-state friction laws and similar frameworks address the history effect by introducing state variables. At the asperity scale, wear debris can play a critical role and change the friction law, i.e., the underlying state variable. In short, wear particles can be active (supporting load) or inactive (down in valleys), which may cause them to act as lubricants. Understanding their complex influence can be numerically addressed with Discrete-Element Methods (DEM). We present a hierarchical multiscale framework that links particle (meso-)scale dynamics to a (macro-) Finite-Element model (1D), where the friction law is governed by particle density and Archard’s law. Discrete-element and boundary-element simulations are coupled, allowing multiple oblate-spheroidal debris particles to evolve between rough surfaces. In detail, particles trapped between surfaces will 1) redistribute normal loads at the interface and 2) lower tangential sliding resistance. The evolution of active particle's density, aka. the fundamental state for the friction law, is then obtained in relation with normal pressure, sliding velocity, and surface topography. The macroscopic friction law is validated against experimental strip-draw tests, where the wear particle sizes and surface roughness were measured to set up BEM–DEM simulations, and where macroscopic sliding resistance was measured experimentally. Macroscopic conditions (contact area, pressure, sliding velocity) affect the wear process, and we demonstrate our model's ability to predict the influence of normal pressure as well as the tool-pad size on friction.