An AI‐Enabled Integrated Blended Teaching Framework for Civil Engineering Education: A Case Study From an Undergraduate Program in China

ABSTRACT The rapid advancement of artificial intelligence (AI) is reshaping engineering education by promoting more adaptive, data‐informed, and practice‐oriented learning paradigms. To address persistent challenges in undergraduate civil engineering education, including insufficient personalized support, limited integration of authentic engineering practice, and over‐reliance on summative assessment, this study proposes an AI‐enabled integrated blended teaching framework. The framework integrates large language model‐assisted learning support, blended instructional design, authentic engineering practice scenarios, and a multidimensional process‐oriented assessment system. It was implemented in the undergraduate course Principles of Concrete Structure Design, where mainstream large language models, including Wenxin Yiyan, ChatGPT, and Tongyi Qianwen, were integrated with the Rain Classroom platform to support question answering, learning guidance, resource recommendation, and feedback generation. Course materials, engineering codes, and case documents were incorporated to constrain AI responses, while instructor review and weekly verification were adopted to reduce hallucination risks. Dynamic student profiles were constructed based on learning behavior data to support differentiated learning recommendations. Evaluation using platform logs, academic performance, questionnaires, and inferential statistical analyses showed that the framework was associated with improved engagement, more stable participation, better understanding of engineering practice, and more timely instructional feedback. The findings suggest that embedding AI into instructional mechanisms can support personalized learning, engineering‐oriented teaching, and process‐based evaluation in civil engineering education.