research-48

Research Areas

artificial-intelligence-48

Artificial intelligence for safety and health

classroom-48

Education and training

safety-hat-48

Design for Safety (DfS) and safety management

roadblock-48

Fall hazards and controls

artificial-intelligence-48

Artificial intelligence for safety and health

Computer vision for site safety and housekeeping

This project applies computer vision to enhance construction site safety by detecting poor housekeeping, missing barricades, and hazardous conditions through automated image and video analysis. Using data from CCTV, drones, and body-worn cameras, the system performs image classification, object detection, and change detection to identify risks in real time.

Our tools include AI-enabled dashboards, Telegram alerts for site teams, and customisable reports to support timely interventions and improve overall site conditions. We also explored supervised learning approaches to identify cluttered environments and dynamically detect crane load fall zones.

Publications

  1. Chian, E., Fang, W., Goh, Y. M., & Tian, J. (2021). Computer vision approaches for detecting missing barricades. Automation in Construction, 131, 103862. https://doi.org/10.1016/j.autcon.2021.103862
  2. Chian, E. Y. T., Goh, Y. M., Tian, J., & Guo, B. H. W. (2022). Dynamic identification of crane load fall zone: A computer vision approach. Safety Science, 156, 105904. https://doi.org/10.1016/j.ssci.2022.105904
  3. Lim, Y. G., Wu, J., Goh, Y. M., Tian, J., & Gan, V. (2023). Automated classification of “cluttered” construction housekeeping images through supervised and self-supervised feature representation learning. Automation in Construction, 156, 105095. https://doi.org/10.1016/j.autcon.2023.105095
  4. Sun, K., Shao, Z., Goh, Y. M., Tian, J., & Gan, V. J. L. (2025). Change detection network for construction housekeeping using feature fusion and large vision models. Automation in Construction, 172, 106038. https://doi.org/10.1016/j.autcon.2025.106038

Projects

Construction safety leading indicators

This project focuses on identifying and modelling leading indicators to proactively assess subcontractor safety risk and forecast safety performance outcomes. Leveraging safety analytics and machine learning techniques, the research extracts meaningful insights from historical data to support the early identification of high-risk contractor profiles and strengthen safety management practices.

These insights enable organisations to prioritise preventive interventions, track safety performance, and develop a data-informed safety culture that is proactive rather than reactive.

Publications

  1. Poh, C. Q. X., Ubeynarayana, C. U., & Goh, Y. M. (2018). Safety leading indicators for construction sites: A machine learning approach. Automation in Construction, 93, 375–386. https://doi.org/10.1016/j.autcon.2018.03.022
  2. Kam, S. H., Lan, T., Sun, K., & Goh, Y. M. (2025). Feature weights in contractor safety performance assessment: Comparative study of expert-driven and analytics-based approaches. Automation in Construction, 174, 106142. https://doi.org/10.1016/j.autcon.2025.106142
  3. Sun, K., Lan, T., Goh, Y. M., & Huang, Y.-H. (2024). Overcoming imbalanced safety data using extended accident triangle. arXiv. https://doi.org/10.48550/ARXIV.2408.07094

Projects

Text mining

This project applies machine learning and natural language processing (NLP) techniques to automatically classify construction accident descriptions. By leveraging text mining, organisations can extract actionable insights from unstructured safety reports and documentation, enhancing their ability to identify trends, root causes, and preventative measures. These insights support data-driven decision-making to reduce accidents and improve workplace health and safety.

Publications

  1. Goh, Y. M., & Ubeynarayana, C. U. (2017). Construction accident narrative classification: An evaluation of text mining techniques. Accident Analysis & Prevention, 108, 122–130. https://doi.org/10.1016/j.aap.2017.08.026
classroom-48

Education and training

Digital simulation games

This project involves the design and development of 3D digital game-based learning (DGBL) tools to enhance construction safety training and education. These interactive simulation games immerse learners in realistic scenarios, helping them develop critical safety competencies in a risk-free environment.

SafeSim Risk (SSR) is a multiplayer game designed to train players in hazard identification and safe work practices on a virtual construction site. SafeSim Investigation (SSI) teaches players how to conduct incident investigations by collecting digital evidence and uncovering root causes through gameplay. SafeSim Design (SSD) is a DGBL platform created specifically for designers to learn about construction design risks and effective mitigation strategies. Its content is aligned with the IES-NUS DfS Library to reinforce upstream safety principles. These games aim to cultivate hazard awareness, investigation skills, and design-for-safety thinking through engaging, experiential learning.

ProjectSim is designed to enhance project management competencies through immersive, scenario-based learning. By navigating realistic project challenges, learners can apply key techniques such as earned value management (EVM) and earned duration management (EDM) to monitor performance, manage risks, and make data-driven decisions. 

Publications

  1. Lee, Y. Y. R., Samad, H., & Miang Goh, Y. (2020). Perceived importance of authentic learning factors in designing construction safety simulation game-based assignment: Random Forest approach. Journal of Construction Engineering and Management, 146(3), 04020002. https://doi.org/10.1061/(ASCE)CO.1943-7862.0001779
  2. Safiena, S., & Goh, Y. M. (2022). A hazard identification digital simulation game developed based on the extended authentic learning framework. Journal of Engineering Education, 111(3), 642–664. https://doi.org/10.1002/jee.20459
  3. Safiena, S., & Goh, Y. M. (2024). Authentic learning questionnaire for digital simulation games in higher education: A construction safety case study. Education and Information Technologies, 29(14), 17915–17941. https://doi.org/10.1007/s10639-024-12543-z
  4. Safiena, S., Tay, J., Miang Goh, Y., & Lim, M. (2023). SafeSim Design: A digital game-based learning approach to address Design for Safety (DfS) competency. In G. Geng, X. Qian, L. H. Poh, & S. D. Pang (Eds.), Proceedings of the 17th East Asian-Pacific Conference on Structural Engineering and Construction, 2022 (pp. 360–372). Springer Nature. https://doi.org/10.1007/978-981-19-7331-4_29
  5. Tay, J., Goh, Y. M., Safiena, S., & Bound, H. (2022). Designing digital game-based learning for professional upskilling: A systematic literature review. Computers & Education, 184, 104518. https://doi.org/10.1016/j.compedu.2022.104518

Projects

Immersive technology for learning

This project leverages virtual reality (VR), augmented reality (AR), and mixed reality (MR) technologies to enhance construction and engineering education. We developed a VR and MR-based learning tool for incident investigation, featured in RESCUE Act 3, providing immersive and interactive training experiences. In addition, we created a mobile app that uses AR and VR to teach structural systems to non-engineering students. These tools aim to improve learner engagement and understanding of complex concepts.

Publications

  1. Hu, X., Goh, Y. M., & Lin, A. (2021). Educational impact of an Augmented Reality (AR) application for teaching structural systems to non-engineering students. Advanced Engineering Informatics, 50, 101436. https://doi.org/10.1016/j.aei.2021.101436
  2. Hu, X., Safiena, S., Goh, Y. M., & Lin, A. (2023). Using virtual reality (VR) to improve structural systems knowledge of project and facilities management students. Educational Technology Research and Development, 71(5), 1993–2019. https://doi.org/10.1007/s11423-023-10251-y
  3. Yang, F., & Miang Goh, Y. (2022). VR and MR technology for safety management education: An authentic learning approach. Safety Science, 148, 105645. https://doi.org/10.1016/j.ssci.2021.105645

Projects

Enhancing professional development for adult learners

Our research focuses on advancing lifelong learning though innovative, technology-enhanced strategies tailored for adult learners. We have developed an adaptive learning system to support personalised upskilling for construction professionals. We are currently modelling self-regulated behaviours in video- and game-based learning environments. These projects aim to foster learner autonomy, optimise engagement, and improve performance outcomes in professional education.

Publications

  1. Hu, X., Goh, Y. M., & Tay, J. (2024). Construction professionals’ perspectives of adaptive learning adoption: An SEM-machine learning approach. Engineering, Construction and Architectural Management. https://doi.org/10.1108/ECAM-07-2024-0896

Projects

safety-hat-48

Design for Safety (DfS) and safety management

DfS climate

Design for Safety (DfS) is an upstream approach that brings project stakeholders together early in the design process to identify and mitigate foreseeable risks across the project lifecycle. This project focuses on developing and measuring DfS climate where the shared perceptions among DfS review team members regarding the effectiveness of DfS policies and procedures.

Publications

  1. Lim, M. S., & Goh, Y. M. (2023). Development and validation of the Design for Safety (DfS) climate measurement tool. Journal of Risk Research, 26(12), 1331–1352. https://doi.org/10.1080/13669877.2023.2288003
  2. Lim, M. S. H., Tang, Y., Du, S., & Goh, Y. M. (2025). Beyond compliance: A two-axis model of Design for Safety implementation in mandatory contexts. Journal of Management in Engineering, 41(5), 04025029. https://doi.org/10.1061/JMENEA.MEENG-6492

Safety climate and culture

This project explores how truck drivers and migrant construction workers perceive safety policies, leadership, and practices in their work environments. Through surveys, interviews, and data-driven methods including interpretable clustering, we examine the factors shaping safety attitudes, identify challenges to building a strong safety culture, and propose targeted strategies for improvement. A key focus is understanding how national culture influences safety culture, especially in Singapore’s diverse migrant workforce. By analysing cultural dimensions and their impact on safety behaviours, we aim to develop tailored behavioural interventions that improve safety outcomes and better protect vulnerable worker groups.

Publications

  1. He, Y., Huang, Y.-H., Lee, J., Lytle, B., Asmone, A. S., & Goh, Y. M. (2022). A mixed-methods approach to examining safety climate among truck drivers. Accident Analysis & Prevention, 164, 106458. https://doi.org/10.1016/j.aap.2021.106458
  2. Sun, K., Lan, T., Goh, Y. M., Safiena, S., Huang, Y.-H., Lytle, B., & He, Y. (2024). An interpretable clustering approach to safety climate analysis: Examining driver group distinctions. Accident Analysis & Prevention, 196, 107420. https://doi.org/10.1016/j.aap.2023.107420
roadblock-48

Fall hazards and control

Fall hazard controls

We developed a mobile-based design support tool to assist in selecting and designing fall arrest and travel restraint systems for building construction. The prototype was validated through drop tests in collaboration with BMS Training Specialist and Tractel Singapore.

Publications

  1. Lim, W. C., Tashrif, S. M., Goh, Y. M., & Adrian Koh, S. J. (2021). Validation of the energy balance approach for design of vertical lifeline systems. International Journal of Occupational Safety and Ergonomics, 27(3), 673–685. https://doi.org/10.1080/10803548.2019.1616948
  2. Tashrif, S. M., Lim, W. C., Goh, Y. M., Hu, X., & Koh, S. J. A. (2022). Experimental validation of an energy balance approach for design of horizontal lifeline systems. International Journal of Occupational Safety and Ergonomics, 28(1), 275–288. https://doi.org/10.1080/10803548.2020.1763031

Projects