Abstract: In response to the complex and dynamic construction process of drilling-and-blasting method tunnels and the difficulty of meeting the refined requirements of intelligent railway construction with traditional manual recording and experience-based management models, this paper developed a digital twin modeling technology for full-face tunneling using the drilling-and-blasting method, and integrated ultra-wideband (UWB) positioning and machine vision technologies to realize intelligent recognition of construction machinery position and operational status during the construction process. On this basis, the technology enables dynamic collection and quantitative analysis of the timeliness of key processes, and achieves intelligent recognition of five typical processes including drilling, blasting, mucking, arch erection, and shotcreting. Field application results show that the model can significantly improve the efficiency of process handover, reduce construction risks, and provide key technical support for the in-depth advancement of intelligent railway construction.