Urban Energy Modeling (UEM) provides a comprehensive approach to urban planning, helping to create sustainable, resilient, and energy-efficient cities that meet the needs of current and future generations. The key inputs for UEM methodologies and tools are the geometry of the building stock and its thermophysical properties. In the Netherlands, the 3DBAG provides the building stock geometry, while the thermophysical properties can be approximated using energy consumption estimates specific to each residential building type from the IEE project TABULA. However, a challenge arises as open data on residential building types at a national level is currently not readily available, necessitating the development of a method to infer this information from other accessible data sources.

In response to similar successful studies, this thesis also focuses on utilising machine learning to address this challenge. Support Vector Machine (SVM) and Random Forest (RF) algorithms were tested and compared. These algorithms were trained using data on the residential building types obtained from the Rijssen-Holten energy testbed and EP-online, with the latter requiring preprocessing to obtain the relevant information. Additionally, 25 features derived from cadastral data and building geometry underwent a selection process to identify the essential features for accurate classification of building types.

Eight models were trained and applied across eight case studies, containing subsets of the Netherlands representative for the whole country. The combined results were analyzed to determine necessary features, required data, and the most suitable machine learning approach for this research.

The results revealed that features such as adjacency to other buildings, width, and volume in LoD2.2 correlated the most with Dutch residential building types. However, critical features like the number of storeys, the presence of an open porch, or galleries were not available as open data, even though they directly relate to the definition of certain residential building types.

This thesis presented successful models, demonstrating accuracies between 61.1% and 98.5%, and balanced accuracies ranging from 51.6% to 94.2%. Importantly, performance differences were observed in various case studies, particularly in distinguishing accuracy between multi-dwelling and single-dwelling houses. Despite the longer tuning and training time, the suitability and accuracy of the RF models generally outperformed the SVM models.

These findings highlights the capacity of machine learning to attain robust classification outcomes for the Dutch building stock when trained on representative datasets. Nevertheless, it is emphasized that the accuracy of results is contingent on data quality, and difficulties may arise in scenarios involving intricate buildings with multiple components and ambiguous classification rules.

A 3D city model uses three-dimensional geometries to represent and model urban environments, in which the building model is the key feature. With the development of computer and data collection technologies, 3D city models are gaining growing capacities regarding storing rich information. This makes 3D city models more potentially useful than ever in the urban application domain.

When implementing 3D city models, the CityGML model is currently the most frequently used standard. It is being used by cities all over the world. CityJSON is an encoding for a subset of the OGC CityGML data model. It is a JSON-based data exchange format for digital 3D models of cities and landscapes, and it is easy for various operations. Due to the uniqueness of its structure, it is necessary to design corresponding data models to store CityJSON files in the database, to make querying and updating data within the database easy and convenient.

There are already some open-source solutions for storing CityJSON files in the database, 3DCityDB is one of them. 3DCityDB can store, manage and visualize data well, and it is open source. But the database design is very complex: for a tile of 3D BAG data, 3DcityDB uses a total of 66 tables to store data. The structure of the data model results in difficulties for database users to understand the imported data, and potentially leads to non-optimal operation performance when retrieving data for urban applications. Based on the drawbacks of the existing DBMS when dealing with the CityJSON data format, this project aims to develop a Postgres data model that can store CityJSON files simply and efficiently. The developed Postgres data model (CJDB) has a simpler table structure and data model design, a CityJSON data importer, and an interactive API user interface.

After going through this document (data model, importer and API section), the potential users will have the ability to:

• Import CityJSON files into a Postgres database,
• Perform queries on imported data,
• Perform operations using CJDB API.

In addition, by reading the benchmarking section, the users can gain an overview of the CJDB’s performance over the 3DcityDB’s.

The CJDB project is open-sourced, available on GitHub page.

The CJDB project will be potentially further developed by 3D geoinformation group of TU Delft and 3DGI.