A vehicle's chassis plays a critical role in its reliability and durability. To ensure occupant safety and vehicle maneuverability, it is necessary to assess the fatigue strength of chassis components, i.e. their ability to withstand repeated loads during use. This assessment begins at the design stage, with the identification of operating conditions and associated loads. In the case of personal vehicles, various loads must be considered due to diverse road types (e.g., highway, city,. . .) and driving styles (aggressive, sporty,. . .). In this paper we use a multi-dimensional characterization of the damage caused by external multi-input loads on wheels during vehicle use, including load combinations between the left and right wheels of the front and rear axles. Field measurements are used to calculate pseudo-damages for each load and road type, creating multivariate data with hierarchical structure. Unsupervised statistical analyses are used to explore correlations between pseudo-damages and identify driving profiles, providing a multi-dimensional assessment of severity while avoiding overlearning. A multi-dimensional Gaussian mixture model is then fitted to damage-equivalent constraints. This probabilistic model extrapolates damage computing and simulates driving styles, providing design teams with a stress analysis tool for accurate, realistic fatigue design of chassis components in future vehicles.