Abstract: Over the past several decades, the abundant and comprehensive data accumulated in the civil engineering field has laid a solid foundation for the application and development of artificial intelligence techniques such as machine learning and deep learning. Most applications have primarily followed a "plug-and-play" approach, lacking understanding and fusion with the unique knowledge systems of civil engineering. Various constraints in applicability and application scope caused by aforementioned issues have increasingly drawn attention from researchers, leading to the emergence of domain knowledge embedded intelligent analysis methods as a new research focus. To systematically sort out and summarize the research progress and state-of-the-art technologies in this field, this study provides a comprehensive review of domain knowledge embedded intelligent analysis methods in civil engineering. Bibliometric analysis software is used to analysis relevant publications in terms of publication time, of institutions, and of keyword co-occurrence, of highlighting the research achievements and of hotspots in domain knowledge fusion methods. This is followed by a systematic abstract and analysis of four major domain knowledge fusion methods: physics-guided, physics-enhanced, physics-corrected, and physics-constrained approaches. The challenges and future directions for domain knowledge embedded intelligent analysis methods are discussed.