Dr. Jeelani leads the Construction Automation and safety (CAS) research group, at the Rinker School of Construction Management. His research focuses on construction safety, visual data analytics, and cognitive sciences to support the building of the next generation of safe and smart infrastructure.
Industry Expertise (4)
Construction - Commercial
Construction - Residential
Areas of Expertise (4)
Improving Safety Performance in Construction Using Eye-Tracking, Visual Data Analytics, and Virtual RealityConstruction Research Congress 2020: Safety, Workforce, and Education
Idris Jeelani, Alex Albert, Kevin Han
2020 Globally, construction is among the most dangerous industries. Among others, research has demonstrated that construction workers and professionals fail to recognize and manage an unacceptable number of safety hazards. To address this, past research has focused on examining several factors including training, education, and management support that may indirectly influence hazard identification and management performance. However, proximal factors associated with poor hazard identification and management at the work interface has received very little attention. This paper summarizes our research aimed at (1) understanding why workers fail to identify and manage safety hazards as they participate in hazard recognition and management efforts and; (2) developing evidence-based interventions to counter poor hazard identification and management performance in the construction industry. This includes examining hazard recognition as a visual search task and understanding how search pattern (i.e., how workers examine the work environment) affects hazard recognition performance. More specifically, the research used eye-tracking technology to examine the relationship between how workers examine the workplace and the resulting hazard identification performance. Based on this new knowledge generated, new interventions were developed to improve hazard recognition and management performance. These include two interventions. First, an immersive hyper-realistic mixed reality training environment that was developed using stereoscopic visual data captured from real construction workplaces. The testing of the intervention with 56 participants suggested that the intervention can significantly improve hazard identification and management performance. Second, an AI-based system that detects hazardous conditions and objects in real-time to assist workers and managers. The system uses the live video captured by a wearable camera to localize the workers on a pre-built global map, detect any hazard present around the worker, and warns them in real-time. The system is tested in indoor and outdoor construction environments, which indicate 93% accuracy in detecting workers’ proximity to hazards.
Developing hazard recognition skill among the next-generation of construction professionalsConstruction Management and Economics
Alex Albert, Idris Jeelani, Kevin Han
2020 Globally, a large number of safety hazards remain unrecognised in construction workplaces. These unrecognised safety hazards are also likely to remain unmanaged and can potentially cascade into unexpected safety incidents. Therefore, the development of hazards recognition skill – particularly among the next-generation of construction professionals – is vital for injury prevention and safe work-operations. To foster the development of such skill, the current investigation examined the effect of administering a hazard recognition intervention to students seeking to enter the construction workforce. First, prior to introducing the intervention, the pre-intervention hazard recognition skill of the participating students was measured. Next, the intervention that included a number of programme elements was introduced. The programme elements included (1) visual cues to promote systematic hazard recognition, (2) personalised hazard recognition performance feedback, (3) visual demonstration of common hazard recognition search weaknesses, and (4) diagnosis of hazard search weaknesses using metacognitive prompts. Finally, the post-intervention skill demonstrated by the student participants was measured and compared against their pre-intervention performance. The results suggest that the intervention was effective in improving the hazard recognition skill demonstrated by the next-generation of construction professionals. The observed effect was particularly prominent among those that demonstrated relatively lower levels of skill in the pre-intervention phase. The research also unveiled particular impediments to hazards recognition that the participants experienced.
Development of virtual reality and stereo-panoramic environments for construction safety trainingEngineering, Construction and Architectural Management
Idris Jeelani, Kevin Han, Alex Albert
2020 Purpose Workers and construction professionals are generally not proficient in recognizing and managing safety hazards. Although valuable, traditional training experiences have not sufficiently addressed the issue of poor hazard recognition and management in construction. Since hazard recognition and management are cognitive skills that depend on attention, visual examination and decision-making, performance assessment and feedback in an environment that is realistic and representative of actual working conditions are important. The purpose of this paper is to propose a personalized safety training protocol that is delivered using robust, realistic and immersive environments. Design/methodology/approach Two types of virtual environments were developed: (1) Stereo-panoramic environments using real construction scenes that were used to evaluate the performance of trainees accurately and (2) A virtual construction site, which was used to deliver various elements of instructional training. A training protocol was then designed that was aimed at improving the hazard recognition and management performance of trainees. It was delivered using the developed virtual environments. The effectiveness of the training protocol was experimentally tested with 53 participants using a before–after study. [...]
Real-time vision-based worker localization & hazard detection for constructionAutomation in Construction
Idris Jeelani, Khashayar Asadi, Hariharan Ramshankar, Kevin Han, Alex Albert
2020 Despite training, construction workers often fail to recognize a significant proportion of hazards in construction environments. Therefore, there is a need for developing technology that assists workers and safety managers in identifying hazards in complex and dynamic construction environments. This study develops a framework for an automated system that detects hazardous conditions and objects in real-time to assist workers and managers. The framework consists of three independent pipelines for localization of workers, semantic segmentation of the visual scene around workers, and detection of static and dynamic hazards. The framework can be used to automate and augment the hazard detection ability of workers and safety managers in construction workplaces. In addition, the framework offers several computing contributions including an improved real-time worker localization method and an efficient architecture for integrating pipelines for entity localization and object detection. A system developed based on the proposed framework as a proof of concept and was tested in indoor and outdoor construction environments. It achieved over 93% accuracy.
Real-Time Hazard Proximity Detection—Localization of Workers Using Visual DataComputing in Civil Engineering 2019: Data, Sensing, and Analytics
Idris Jeelani, Hariharan Ramshankar, Kevin Han, Alex Albert, Khashayar Asadi
2019 Research indicates that workers often fail to recognize a significant proportion of safety hazards. To reduce injury likelihood, efforts have traditionally focused on developing and delivering training interventions. Despite such efforts, desirable levels of hazard recognition are rarely achieved. Therefore, augmenting human abilities with a technology-driven solution to improve overall hazard recognition can yield substantial benefits. Accordingly, the objective of this study is to develop a method for localizing workers with respect to pre-identified hazards in real-time. To achieve this objective, a 3D point cloud of a construction site as a global map is created and hazard locations are marked on this map. Workers are provided with a head-mounted camera that continuously records their first-person view (FPV) videos. The image frames from these videos are localized onto the global map using bag of word (BoW) localization. Apart from estimating the proximity to safety hazards, the system can also capture large-scale data that captures unsafe behaviors (e.g., entry to restricted areas) and near-miss incidents for training purposes.
- Construction Automation and Safety (CAS) : Leader