Anti-optimization based uncertainty analysis for composite laminates

Composite materials are used in diverse engineering applications and are widely applied in high-performance tasks such as in  aircraft, automobiles and in the military industry. Over the past few decades, many researchers studied the behavior of these materials, promoting big advances in the area. Although great progress has been achieved, variability from experimental data obtained for, a priory, and nominally identical specimens is still not well explained. Random and epistemic uncertainties are often pointed as the main causes of this divergence. This work presents an algorithm based on interval analysis to quantify uncertainties in composite laminates. The α-cut procedures are used to take into account various uncertainties present in the material, geometry, or external loading. The solutions found in this work are compared with the robust Monte Carlo method, showing that the algorithm presented here finds wider intervals than those found by Monte Carlo, indicating a more precise analysis.