| Project Reference: | ITP/018/20LP |
| Project Title: | A Computer Vision-enabled Digital Twin for Construction Resource and Progress Management |
| Hosting Institution: | The Hong Kong Polytechnic University (PolyU) |
| Abstract: | The construction industry has played a key role in Hong Kong and been an instrumental driver of economic growth and enabler of social development for many generations. Whilst the construction industry has a long and proud history, it has not been without its challenges, among which high costs and declining productivity have been highlighted in recent years. This has led to adverse effects on the broader Hong Kong economy and community. The conventional construction management relies on sampling labour productivity data to form a baseline to control the overall project budget and progress. Although there are numerous diverse activities on sites, site management can only manually inspect (sample) a limited number of activities to gauge the overall cost and productivity. This error-prone process inevitably leads to mis-judgment and unnecessary resource overrun, and post-event remedies (rework). The industry is longing for a new management tool for better construction project management. In this project, we will develop a Digital Twin technology to support construction management. Digital Twin is a concept of having a real-time digital representation of a physical object. Usually, digital data is formed by sensors that continuously monitor changes in the environment and report back the updated state in the form of measurements and pictures. Progress monitoring verifies that the completed work is consistent with plans and specifications. A physical site observation is needed in order to verify the reported percentage of work done and determine the stage of the project. The digital twin concept, paired with wearable and mobile devices on a construction site, can help to better represent the as-build project at any point in time. By an as-built state of a building or structure, we can compare it with an as-planned execution in BIM and take corresponding actions to correct any deviations. The digital twin allows up-to-date information to be fed back to the field so as to provides automatic resource allocation monitoring and waste tracking. The technology decreases the number of errors and reworks; and allowing for a predictive and lean approach to resource management. In the previous project ITF project (ITP/020/18LP), the project team has implemented and test computer vision (CV) technologies in enhancing construction quality management. With the initial success, we propose to extend the application of CV technologies to construction resource and progress management. This project will develop a computer vision-enabled digital twin based on far, middle, and near surveillance videos. Five typical trades will be investigated: formwork, reinforcement, concrete placing, plastering, and tiling. Furthermore, four features will characterize the digital twin: 1) comprehensive data collection and statistics, 2) automatic analysis and comparison, 3) proactive project risk diagnosis, and 4) learning-based resolution recommendation. This study will contribute to transforming the construction management practice to be a population-based, intensive, and proactive process; it will improve resource productivity, reduce resource waste and progress lagging, as well as improve project performance. Additionally, the project has the potential to improve the competitiveness of the Hong Kong construction industry. The success of this project could also promote creativity and innovation. Multiple organizations, including Chun Wo Building Construction, Yau Lee Engineering, A&D Engineering, etc., have all expressed strong interests in providing cash donations and conducting trials for the technologies. We believe that there is a large demand for this technology and the results of the project will have a large impact on the construction industry. |
| Project Coordinator: | Prof Heng Li |
| Approved Funding Amount: | HK$7.20M |
| Project Period: | 01 Dec 2020 - 30 Nov 2022 |
