Airborne laser technology produces point clouds that can be used to build 3D models of buildings. However, the work is a laborious process that could benefit from automation. Artificial intelligence (AI) has been widely used in automating building segmentation as one of the initial stages in the 3D modeling process. The algorithms with a high success rate using point clouds for automatic semantic segmentation are random forest (RF) and PointNet++, with each algorithm having its own advantages and disadvantages. However, the training and testing data to develop and test the model usually share similar characteristics. Moreover, producing a good automation model requires a lot of training data, which may become an issue for users with a small amount of training data (limited data). The aim of this research is to test the performance of the RF and PointNet++ models in different regions with limited training and testing data. We found that the RF model developed from a small amount data, in different regions between the training and testing data, performs well compared to PointNet++, yielding an OA score of 73.01% for the RF model. Furthermore, several scenarios have been used in this research to explore the capabilities of RF in several cases.

Before the development of 3D cadastre, cadastral systems were based on 2D representations, which now require transformation or updating. In this context, the first issue is that existing 2D rights are not aligned with recent 3D data acquired using advanced technologies such as Unmanned Aerial Vehicle–Light Detection and Ranging (UAV-LiDAR). The second issue is that point clouds of objects captured by UAV-LiDAR, such as fences and exterior building walls—are often neglected. However, these point cloud objects can be utilized to adjust 2D rights to correspond with recent 3D data and to update 3D building models with a higher level of detail. This research leverages such point cloud objects to automatically generate 3D rights and building models. By combining several algorithms, such as Iterative Closest Point (ICP), Random Forest (RF), Gaussian Mixture Model (GMM), Region Growing, the Polyfit method, and the orthogonality concept—an automatic workflow for generating 3D cadastral models is developed. The proposed workflow improves the horizontal accuracy of the updated 2D parcels from 1.19 m to 0.612 m. The floor area of the 3D models improves by approximately ±3 m². Furthermore, the resulting 3D building models provide approximately 43% to 57% of the elements required for 3D property valuation. The case study of this research is in Indonesia.

Land taxation, mainly the Land and Building Tax (LBT), is a critical revenue source for local governments in Indonesia, yet its contribution remains suboptimal due to inefficiencies in assessment and administration. This study focuses on the Lebak Regency to optimize the Sales Value of Taxable Objects (SVTO) assessment, a key determinant of LBT revenue. Using spatial analysis, Land Value Zones (LVZ), and socio-economic data, the research evaluates assessment ratios to identify discrepancies between SVTO and market values. Adjustments to SVTO are proposed based on spatial patterns, such as land use and infrastructure, and the Index of Developing Villages (IDV). Findings reveal significant disparities between assessed and market values, with assessment ratios ranging from 1% to 267%. Simulations indicate potential LBT revenue of IDR 224.7 billion under full compliance, compared to IDR 44.22 billion, the 2024 target. Adjustments based on spatial patterns from land use planning and village development indices enhance equity and accuracy in tax assessments, optimizing local revenues. Despite these improvements, the study’s limitations lie in the lack of community validation for the proposed methodology, which is essential to confirm its practicality and acceptance. Future research should address this gap and explore household-level dynamics, including tax affordability and spending patterns, to enhance policy inclusivity and align taxation systems with local socio-economic realities.

Parcels data in Indonesia are still stored in 2-dimensional (2D) geometry which are integrated with other attribute data, such as the case with the Directorate of Land and Building Tax Indonesia. Whereas, building taxes assessment refers to a number of details that require the information to be stored in 3-dimensional (3D) forms. This study aims at the use of Building Information Modeling (BIM) technology, which widely used in building asset management in 3D. This research illustrates the usability of the role of BIM in assessing and managing building taxes in Indonesia. The point clouds were obtained using Terrestrial Laser Scanner (TLS) technology. The point clouds processed so that it can form 3-dimensional geometrical apartment. The attributes of the 3-dimensional model integrated with the geometric model using the BIM concept. The results analyzed to assess whether BIM concept was able to fulfill the needs of the 3D fiscal cadastre in Indonesia.

Property taxation serves as a critical instrument for fiscal efficiency and equitable distribution, yet implementation faces significant challenges including valuation inaccuracies, insufficient administrative capacity, and diminished public trust. Indonesia’s Land and Building Tax (PBB-P2) utilizes the Sales Value of Taxable Objects (NJOP) as an administrative proxy for market value, which frequently deviates from actual land prices. These disparities create horizontal inequities, diminish local revenue potential, and generate taxpayer resistance, especially in decentralized regions with constrained technical resources. This research presents the Spatially Adaptive Land Value Determination (SALAD) model as a comprehensive framework for enhancing property tax governance and modernizing fiscal cadastre systems. Unlike conventional mass appraisal methods, SALAD integrates spatial zoning, assessment ratio analysis, land-use characteristics, and the Index of Developing Villages (IDM) with socio-economic indicators including purchasing power and community fiscal behavior. The model incorporates structured social validation to improve public acceptance. Field validation in Lebak Regency employed mixed-methods design with surveys of 75 respondents across 20 villages and interviews with village heads and tax officials. Results demonstrate that transparency, fairness, and visible public benefits are essential for community support. Validation indices vary significantly by IDM category (ANOVA: F = 4.23, p = 0.03 for economic; F = 3.81, p = 0.04 for institutional), confirming that the SALAD model’s adaptive mechanism is empirically grounded.

Building Information Modeling (BIM) technology has been widely adopted in the construction industry for managing building information. However, its application in cultural heritage conservation has yet to be fully explored. This study aims to demonstrate the potential of using BIM technology in managing cultural heritage tourism assets through a case study of the traditional village of Penglipuran in Bali, Indonesia. Four queries were developed using Industry Foundation Classes (IFC) data model to retrieve information on cultural, legal spaces, physical boundaries, and traditional house elements. The results show that BIM technology can effectively assist in managing cultural heritage tourism assets by providing detailed IFC information. Moreover, results also highlight the potential of BIM technology in preserving and promoting cultural heritage tourism assets. This research provides a comprehensive understanding of BIM technology's potential in cultural heritage conservation and its application in tourism.