Integrating AI-Based Reinforcement Learning and Competency-Based Assessment for Skill and Competency Development

Engineering graduates must be prepared to solve future, undefined problems. To develop necessary competencies, innovative methods like AI-based reinforcement learning (RL) and competency-based assessments are essential. A new AI-RL framework integrates the affective domain into the teaching-learning process using a state-action student database and the SAM model. Based on data from 300 students across 36 subjects (12,960 samples), this approach showed a 7.7% improvement in learning outcomes (p = 0.0041). The scope includes designing AI-based assessment tools, implementing the RL framework, integrating affective domain metrics, and validating improvements in competency development across all learning domains. It also involves continuous data collection and analysis to refine the framework and developing faculty training modules for effective deployment. Scalability across institutions and disciplines will be explored for broader impact. Student details, including basic information, subjects, and section, must be collected and mapped to the SAM process software model. Subject teachers for each section will continuously assess students across all learning domains. The AI-based reinforcement learning (RL) method is then applied to identify individual learning patterns based on each student’s current state, enabling subject matter experts to conduct ongoing assessments. Based on this analysis, targeted actions are recommended to help students develop competencies in specific domains. These action plans are aligned with Bloom’s Taxonomy and categorized under content knowledge, core competencies, critical thinking, core values, and managerial skills, all implemented through the RL framework.