Abstract: With the continuous development of sensing and intelligent technology, the operational monitoring of tunnel boring machines is becoming increasingly perfect. The massive measured in-situ data not only record the important information of the operation process of tunnel boring machine but also involve the internal mechanism of tunnel boring machine and the external mechanism with the working environment. It is of great significance in improving the design, analysis, operation, and maintenance level of tunnel boring machine through mining these in-situ data. In order to summarize and analyze the research and application status of in-situ data of tunnel boring machine, the source, composition, and characteristics of in-situ data of tunnel boring machine are discussed first. Then, the domestic and foreign literature are reviewed from three aspects: data-driven state recognition and performance prediction of tunnel boring machine, data-driven geology recognition and ground surface change prediction and tunnel health monitoring and early warning. The difficulties, advantages, and deficiencies of current researches are discussed as well. Finally, the preliminary analysis and prospects on the future research directions are made from the aspects of in-situ data preprocessing methods, heterogeneous in-situ data modeling methods, generalization ability improvement methods of in-situ model, computing platform, and so on. It is expected to provide inspiration and references for the follow-up big data studies of tunnel boring machine.