Through the forced intrusion of aqueous liquids into hydrophobic zeolitic imidazolate frameworks (ZIFs), shock absorbers can be designed that are both reusable and exhibit high absorption efficiencies, making them attractive for a variety of applications. The intrusion and extrusion processes, and associated absorption efficiency, highly depend on the structural characteristics of the ZIF and the liquid composition, both of which are highly tuneable. Yet, a fundamental understanding of how and why these choices alter the nanoscale intrusion mechanism and impact the intrusion and extrusion processes in macroscopic ZIF crystals remains lacking, preventing their rational design. In this project, I aim to provide such in-depth insight by developing a hierarchical kinetic Monte Carlo model that systematically accounts for different complexities in the {ZIF+liquid} system and by comparing my predicted in silico results with new experiments. My model will be parametrised at near-quantum accuracy through three key nanoscale parameters - informed by my proof-of-concept results – which will be determined based on atomistic simulations with a ZIF-transferable machine-learning interatomic potential. By accounting for the role of different ZIF structures, different liquids, and the ZIF surface, I will unveil the causal impact of each of these choices on the intrusion and extrusion mechanisms, thereby revealing design rules to develop highly innovative shock absorption systems.