A computationally-integrated simulation-focused problem-based learning framework for engineering dynamics: design, implementation, and mixed-methods evaluation in an OBE-aligned undergraduate course

Abstract

Engineering dynamics courses demand simultaneous development of mathematical modelling, computational implementation, and communication competencies, yet conventional instruction rarely provides authentic contexts for their integration. The primary objective of this study was to design, implement, and empirically evaluate a Simulation-Focused Problem-Based Learning (SF-PBL) framework that integrates first-principles Newtonian and Lagrangian mechanics with Python-based numerical simulation within an Outcome-Based Education (OBE)-aligned curriculum. The framework was embedded within a 15-week undergraduate Dynamic System Modeling (DSM) course ( N = 53, B.Tech Mechatronics Engineering, Semester IV, January–April 2026) at a single NBA-accredited institution in India. It operationalises a four-phase computational pipeline—System Identification, Mathematical Modelling, Computational Simulation, Validation & Communication—scaffolded through an Explain-Try-Challenge (E-T-C) model and assessed via a six-judge external expert panel. A pre–post mixed-methods design measured self-efficacy (10-item Likert scale), conceptual knowledge (5-item MCQ), PBL satisfaction, AI tool usage, and qualitative breakthrough reflections. Self-efficacy improved significantly (pre: M = 2.92, SD = 0.21; post: M = 3.80, SD = 0.40; t (52) = 13.91, p < .001 , Cohen’s d = 2.72, pooled SD). Knowledge gains were large ( t (52) = 11.32, p < .001 , d = 1.20; Hake’s g = 0.50, medium interactive-engagement band). All ten self-efficacy items improved significantly ( p < .001 ). PBL satisfaction was high ( M = 3.74 , SD = 0.30). Eighty-five percent of students used AI tools, primarily for computational coding, while mathematical derivation remained predominantly human-led. Qualitative themes identified simulation validation, PID tuning, and linking derivation to computation as commonly reported breakthrough moments. Exhibition rubric scores from 17 student projects evaluated by the six-judge panel ( M = 82.62 / 100 , SD = 9.98 ) confirmed strong authentic performance. The SF-PBL framework offers a replicable, empirically-evaluated model for computationally-integrated dynamics pedagogy in OBE-aligned STEM programmes.