Laminated composite materials play a crucial role in various engineering applications due to their exceptional mechanical properties. This article explores the analysis of reliability and the local sensitivity of failure probability associated with laminated composite materials. It delves into the impact of uncertainties on the performance of these materials and presents an active learning-based reliability methodology. This methodology combines Monte Carlo simulation with a metamodel derived from the combination of three metamodels: artificial neural networks, support vector regression, and Kriging. The article illustrates this methodology through two practical applications on laminated composite plates and compares its performance with other reliability methods. This approach offers valuable insights to enhance the analysis of reliability, strengthen the design process, and facilitate decision-making while fully considering uncertainties related to material properties.