In the context of digital transformation, traditional service design education evaluation models face three core challenges: excessive subjectivity in evaluation criteria, insufficient timeliness of learning feedback, and a disconnect between teaching objectives and evaluation implementation. This study innovatively integrates Outcome-Based Education (OBE) theory with artificial intelligence multimodal analysis technology to construct a dynamic, closed-loop AI-driven evaluation framework. The framework is oriented toward occupational competence requirements and employs a reverse design approach to establish a three-level mapping relationship between "course objectives, learning tasks, and evaluation indicators." It constructs a quantitative indicator system covering design thinking, innovative expression, project collaboration, and reflective abilities. It uses the Analytic Hierarchy Process (AHP) and the Delphi method to determine weight distribution. Technologically, the system employs a three-tier architecture of "data collection—AI analysis—feedback decision-making," facilitating the deep integration and intelligent evaluation of multimodal learning outcomes, including text, images, and speech. It also introduces a human-machine collaborative evaluation mechanism, "AI pre-evaluation—teacher fine-tuning," to balance objective quantification and subjective judgment. Empirical Research demonstrates that the system significantly outperforms traditional evaluation methods in terms of evaluation consistency, achievement of learning objectives, and user satisfaction, showcasing strong theoretical innovation and practical application value.
Gaosheng Luo (2026) studied this question.