EXPLORING AI-EMPOWERED PROJECT-BASED TEACHING REFORM IN MATERIALS MECHANICS: A CASE STUDY OF SNOW LOAD PROBABILISTIC ANALYSIS

Within the undergraduate curriculum of mechanics of materials, a central pedagogical challenge in cultivating top-tier innovative engineering talents lies in guiding students to recognize and comprehend the pervasive uncertainties inherent in real-world engineering problems, while maintaining the relative stability of the course’s knowledge structure and theoretical framework. This study develops an artificial intelligence–empowered project-based instructional framework that systematically integrates stochastic analysis and structural reliability theory into the solution process of specific engineering problems. Employing the first-order reliability method (FORM) as the primary analytical tool, snow load is selected as a representative engineering scenario to establish a load–resistance probabilistic model. A double-iteration algorithm is further implemented to achieve numerical evaluation of structural failure probability. Teaching practice indicates that, without compromising the foundational learning objectives of mechanics of materials, the proposed project-based instructional model effectively enhances students’ modeling competence and engineering decision-making awareness in complex engineering contexts.