DGCNN을 사용한 건설 모니터링의 진척도 정확성 개선

Abstract

Due to the inefficiency of traditional methods, construction monitoring using smart equipment and BIM is gradually increasing due to the need for accurate progress measurement and automatic management. However, there are problems with complex integration and accuracy of element classification using PCD due to the difference in formats used to classify BIM and PCD. Therefore, the purpose of this study is to suggest a method to easily identify the progress by comparing PCD and PCD, increase the accuracy of element classification, and improve construction monitoring. In this study, PCD was created while maintaining model information using IFC and GUID in the design model, and the progress was identified by comparing it with the PCD obtained on site. In addition, for verification, classification was performed according to the schedule and verification was performed using the DGCNN algorithm. Through this, problems of complexity and accuracy are improved, and monitoring using more accurate information is possible by utilizing the obtained information.