Tunnel surveying requires fast and accurate data processing because a tunnel excavation pattern and its work sequences are dependent on the surveying result in the field. While a total station is widely used for measuring reference points in tunnel projects, the data from the total station are too limited in implementing some of the functions required for project control. Recently, 3D laser scanners have been tested in some tunnel projects, because they generate high-density data within several minutes. On the other hand, software development to facilitate processing the resulting large dataset is essential for its wide adoption on tunnel projects. We propose efficient data processing algorithms to accelerate continuous data processing and to facilitate downsizing data points, considering the characteristics of tunnel projects using The New Austrian Tunneling Method.