The demand for sustainable civil and aerospace structures has driven the development of advanced materials and designs. An optimization aimed at minimizing material usage often leads to slender and thin-walled structures, whose load-bearing capacity is constrained by buckling. Fiber-steered composites are a highly promising advanced material concept that can be applied to next-generation composite structures. By employing curved fiber paths, these composites can achieve higher buckling resistance than classical composites. However, uncertainties resulting from manufacturing processes, e.g., variations in fiber orientation angles, gaps, and overlaps, can significantly influence buckling performance. In this work, a polymorphic uncertainty quantification is performed, accounting for both aleatory (inherent variability) and epistemic (lack of knowledge) uncertainties in fiber-steered composites. Based on a previous work [1], fiber paths and geometric imperfections are quantified using fuzzy functions and interval probability-based random fields, respectively. An optimization task considering uncertainties requires a multi-loop algorithm. For this purpose, a multilevel surrogate modeling approach [2] is presented, in which artificial neural networks are trained to predict the buckling behavior with respect to imperfections, as demonstrated in [3]. The results of the multi-objective design optimization with polymorphic uncertain parameters are inherently nested, as described in [4]. Robustness and performance measures are evaluated and visualized via Pareto fronts. The results of the multi-objective optimization of fiber-steered panels are compared with those of classical deterministic buckling design optimizations. References [1] Fina, M. and Bisagni, C., Buckling design optimization of tow-steered composite panels and cylindrical shells considering aleatory and epistemic uncertainties, Comp. Mech. 76, pp. 59–92, 2025. [2] Freitag, S. et al., Multilevel surrogate modeling approach for optimization problems with polymorphic uncertain parameters, Int. J. Approx. Reason., 119, pp. 81-91, 2020. [3] Schweizer, M. et al., Artificial neural networks for random fields to predict the buckling load of geometrically imperfect structures, Comput. Mech. 76, pp. 181–204, 2025. [4] Schietzold, F. N. et al., Robustness versus Performance – Nested Inherence of Objectives in Optimization with Polymorphic Uncertain Parameters, Adv. Eng. Softw., 156, 102932, 2021.