Enhancing Efficiency of Handling Illegal Constructions Using the SOMCM Algorithm: A Case Study in Taipei
Year: 2025
Type: article
Abstract: Illegal constructions (ICs) have raised significant concerns regarding public safety, hygiene, and security in urban areas, posing a continuous risk of accidents and hazards. Whether it be frequent fires in unauthorized buildings or environmental contamination caused by illegal factories, these issues have become crucial for the public and government. In terms of efficiently assigning resources required to manage ICs, responsible authorities find it overwhelming to keep up with the rate of violations when this unwanted situation increases sporadically. To address this issue, the current research proposes a new illegal construction monitoring approach to enhance the efficiency of handling ICs by applying the self-organizing map-based cluster merging (SOMCM) algorithm on a city-scale illegal construction big data. Data collection includes all ICs in Taipei City since 1995, when regulations and the database were updated, resulting in over 370,000 records. Interviews with seven urban construction management experts identified key impact factors for monitoring ICs, verifying five main variables: ICs’ administrative district, position, area, building materials used, and construction progress. Applying the SOMCM algorithm, the study revealed that different administrative districts require tailored handling strategies. Also, most unauthorized constructions are newly built, and roof terraces represent the highest proportion of unauthorized constructions. By spotting violation-prone locations, the insights derived from this study illustrate practical measures employable by relevant authorities when prioritizing and allocating resources for addressing illegal construction cases, thereby minimizing resource waste.
Authors: Jieh-Haur Chen, Mu-Chun Su, Joseph Remir Eklou, Wei-Jen Lin, Po-Han Chen, Nai-Kai Cheng
Language: English
Subfield: Urban Studies
Field: Social Sciences
Domain: Social Sciences
Open Access status: hybrid
OpenAlex ID: W7082162601
