ABSTRACT: The study tackles operational challenges in shield tunnel maintenance, focusing on multi-source date fusion and decision-making suggestion. Employing a multi-disciplinary approach spanning civil engineering, artificial intelligence, and engineering management, the study establishes an intelligent decision-making system based on knowledge graph for shield tunnel maintenance work. Firstly, a multi-data processing program is developed for dealing with multi-source tunnel data, encompassing diverse dimensions reflecting the structural state of shield tunnels. Comprehensive perception of tunnel structure state is achieved through the integration of tunnel internal surface images and 3D point cloud information. Secondly, to address the specialized expert knowledge and experience in tunnel maintenance work, the study explores methods for knowledge representation, extraction, and integration. Leveraging several natural language processing techniques, including ontology and semantic rules. Lastly, a knowledge graph for shield tunnel maintenance is then constructed by Neo4j, integrating data from both multi-source tunnel data and expert knowledge to quantify tunnel structural health and formulate operational strategies for maintenance work. Through the case study, the tunnel knowledge graph is demonstrated to be effective and advanced in tunnel maintenance work. 1. INTRODUCTION AND BACKGROUND In recent years, China's major cities have experienced significant and rapid development in their urban metro systems, aiming to enhance urban transport capacity and address the commuting needs of residents. Zhang et al. (2023) reported, by December 31, 2023, the urban metro network in China extended to 11,232 km, facilitating 23.71 billion passenger trips, resulting in an annual passenger turnover of 197.8 billion trips per kilometer. The growth in urban metro mileage is visually represented in Fig.1. Since 2013, the operational mileage of China's urban metro systems has consistently increased, demonstrating a rising annual growth rate and robust development in domestic urban rail transit. Fig.2. illustrates the total operational rail transit mileage for major cities in 2023, with Shanghai leading with a total of 965 km. Situated on the eastern coast of China, Shanghai, as a developed city, relies significantly on its transportation infrastructure to ensure normal urban operations. The operation of the urban metro service exposes the metro tunnel structure to various risk factors, including periodic train loads, segment aging, and surrounding construction activities.