Key points are not available for this paper at this time.
The expanding number of elderly and disabled populations, coupled with the growing demand for accessible accommodation and the shortcomings of conventional inspection methods, has become a critical global concern. This research emphasises the importance of accessibility in current building codes, focusing on universal design features and their compliance within urban environments. The primary objective was to investigate challenges in conventional and advanced building inspection methods and explore how machine learning (ML) technologies can transform accessibility inspections. Additionally, the study aimed to develop a conceptual framework leveraging OpenCV and ML to detect accessible housing features and ensure compliance with accessibility standards. Key findings from the review include: (1) highlighting inclusive design by addressing the needs of aging and disabled populations, (2) identifying significant limitations in conventional inspection methods, such as inefficiency and subjectivity, (3) emphasising the role of datasets, photogrammetry, and Lidar point cloud data in improving accuracy for accessible design evaluation, and (4) demonstrating how integrating BIM and ML can enhance consistency in compliance verification. The proposed framework offers a systematic approach to automating inspection procedures, showcasing adaptability and scalability for urban housing accessibility assessments. While the framework is yet to be implemented, it provides a strong foundation for certifying accessible housing and offers future directions for real-world applications and empirical studies. These advancements in digital engineering have the potential to transform inspection approaches, supporting accessible housing development on a larger scale and aligning with smart city initiatives to create inclusive urban systems.
Altaf et al. (Thu,) studied this question.