Towards an automated photogrammetry-based approach for monitoring and controlling construction site activities

Omar, Hany

Abstract

The construction industry has a poor productivity record, which was predominantly ascribed to inadequate monitoring of how a project is progressing at any given time. Most available approaches do not offer key stakeholders a shared understanding of project performance in real-time, which as a result failed to identify any project slippage on the original schedule. This study reports on the development of a novel automated system for monitoring, updating and controlling construction site activities in real-time. The proposed system seeks to harness advances in close-range photogrammetry, BIM and computer vision to deliver an original approach that is capable of continuous monitoring of construction activities, with the progress status determinable, at any given time, throughout the construction stage.

The research adopted a sequential mixed approach strategy pursuant to the design science standard processes in three stages. The first stage involved interviews within a focus group setting with seven carefully selected construction professionals. Their answers were analysed and provided "the informed-basis for the development of the automated system” for detecting and notifying delays in construction projects. The second stage involved development of ‘proof of the concept’ in a pilot project case study with nine potential users of the proposed automated system. Face-to-face interviews were conducted to evaluate and verify the effectiveness of the developed prototype, which as a result was continuously refined and improved according to the users’ comments and feedbacks. Within this stage the prototype to be tested and evaluated by a representative of construction professionals was developed. Subsequently a sub-stage of the system’s development sought to test and validate the final version of the system in the context of a real-life construction project in Dubai whereby an online survey is administered to 40 users, a representative sample of potential system users. The third stage addressed the conclusion, limitations and recommendations for further research studies for the proposed system.

The findings of the study revealed that once the system installed and programmed, it does not require any expertise or manual intervention. This is mainly due to all the processes of the system being fully automated and the data collection, interpretations, analysis and notifications are automatically processed without any human intervention. Consequently, human errors and subjectivity are eliminated, and accordingly the system achieved a significantly high level of accuracy, automation and reliability. The system achieved a level of accuracy of 99.97% for horizontal construction elements and exceeded 99.70% for vertical elements. The findings also highlighted that this developed system is inexpensive, easy to operate and its accuracy excels that of current systems sought to automate monitoring and updating of progress status’ for construction projects. The distinctive features of the proposed system assisted the site team to complete the project 61 days ahead of its contractual completion date with a 9% time saving and 3% cost saving.

The proposed system has the potential to identify any deviation from as-planned construction schedules, and prompt actions taken in response to the automatic notification system, which informs decision-makers via emails and SMS.

Citation

Omar, H. Towards an automated photogrammetry-based approach for monitoring and controlling construction site activities. (Thesis). University of the West of England. Retrieved from https://uwe-repository.worktribe.com/output/6000815

• Thesis Type: Thesis
• Deposit Date: May 27, 2020
• Publicly Available Date: Aug 20, 2020
• Public URL: https://uwe-repository.worktribe.com/output/6000815
• Award Date: Aug 20, 2020