**5. Conclusions**

This paper proposes to develop a consolidated 3D data model of underground utilities for land administration. The work includes two parts. On the one hand, a framework for data governance is designed to organize the workflow of utility data survey, management and application through five roles. Through the understanding of current workflow in the utility data usage, this work needs to clearly define the operations and rights of each role in the work process of 3D underground utility mapping. On the other hand, a 3D data model of underground utilities is designed with 3D spatial information, i.e., utility survey information, and land administration information of underground utilities. In order to fill the gap between data capture and usage, this data model has the following main tasks:


A case study is implemented based on the GPR data from the large scale mobile underground utility mapping. The initial implementation transform GPR data from CAD to GIS format and 3D visualization of utilities based on the 3D utility data model. In order to get land administration information, the utility networks have been connecting to the cadastral parcel. The accuracy and details of utility networks need to be improved in future work, such as the spatial relationship between utilities and cadastral parcels. To fully support the land administration of underground space, the 3D utility data model should eventually be extended to include other underground objects and infrastructures in the future, such as underground substations, pedestrian links, common services tunnels, road and rail networks, etc.

This is an ongoing work and in the initial stage. Two main aspects of limitations need to be improved in future work. First, for the accuracy of utility data. Obviously, the GPR data are not enough to provide comprehensive 3D underground utility networks. The other kinds of data (e.g., Gyroscope) should be integrated to provide more precise attributes for underground utilities. Moreover, the details of the shapes and structures of utilities need to be improved. Second, the next step of the data

model development will improve the definition of land administration for underground utilities. Additionally, in order to develop a comprehensive underground utility database, it is necessary to explore the methods to use the existing data and integrate it with newly collected data. The 3D data model should be extended to be 4D (3D + time) to support data updating. A showcase will be developed to realize land administration of underground utility based on a 3D underground utility data model. This will work with a selected agency as data regulatory body and the preferred data integrator. They will help us to evaluate and improve the framework and definition of the data model. After that, recommendations from this showcase will be used to extend the data model include other underground infrastructures and develop the platform of underground space management to support various applications in Singapore.

**Author Contributions:** Conceptualization, J.Y. and K.H.S.; methodology, J.Y. and K.H.S.; implementation, J.Y.; Data curation, J.Y. and S.W.J.; writing—original draft preparation, J.Y.; writing—review and editing, S.W.J., K.H.S., G.S. and A.W.; visualization, J.Y.; supervision, G.S. and A.W.

**Funding:** Underground-related studies and projects fund (USPF), Ministry of National Development and is sponsored by the Singapore Land Authority.

**Acknowledgments:** This publication has been realized as part of the project "Digital Underground: 3D Mapping of Utility Networks" at the Future Cities Laboratory, established by ETH-Zürich and Singapore's National Research Foundation (NRF), and operating under the auspices of the Singapore-ETH Centre.

**Conflicts of Interest:** The authors declare no conflict of interest.
