Search for: All records

Given that roofing contractors in the construction industry have the highest fatality rate among specialty contractors, understanding the root cause of incidents among roofers is critical for improving safety outcomes. This study applied frequency analysis and decision tree data-mining techniques to analyze roofers’ fatal and non-fatal accident reports. The frequency analysis yielded insights into the leading cause of accidents, with fall to a lower level (83%) being the highest, followed by incidence sources relating to structures and surfaces (56%). The most common injuries experienced by roofing contractors were fractures (49%) and concussions (15%), especially for events occurring in residential buildings, maintenance and repair works, small projects (i.e., $50,000 or less), and on Mondays. According to the decision tree analysis, the most important factor for determining the nature of the injury is the nonfragile injured body part, followed by injury caused by coating works. The decision tree also produced decision rules that provide an easy interpretation of the underlying association between the factors leading to incidents. The decision tree models developed in this study can be used to predict the nature of potential injuries for strategically selecting the most effective injury-prevention strategies. 

Recent advances in construction automation increased the need for cooperation between workers and robots, where workers have to face both success and failure in human-robot collaborative work, ultimately affecting their trust in robots. This study simulated a worker-robot bricklaying collaborative task to examine the impacts of blame targets (responsibility attributions) on trust and trust transfer in multi-robots-human interaction. The findings showed that workers’ responsibility attributions to themselves or robots significantly affect their trust in the robot. Further, in a multi-robots-human interaction, observing one robot’s failure to complete the task will affect the trust in the other devices, aka., trust transfer. 

Given the dynamic and complex nature of the construction industry, maintaining situation awareness at job sites is critical. To react properly, workers must identify dynamic safety hazards within the scene. The majority of studies assessing construction workers’ situation awareness have utilized static images, virtual reality, and other types of simulation methods, but questions remain as to whether these formats are able to capture and monitor workers’ naturalistic behaviors and hazard identification abilities. To identify whether the format of hazardous stimuli (i.e., static, image-based vs. dynamic, and video-based formats) impact workers’ subjective and objective hazard identification and situation awareness metrics, this study developed 23 safety hazard scenarios utilizing state-of-the-art augmented 360° panoramas and then tracked differences in workers’ visual search patterns and hazard identification abilities using eye-tracking technology. The workers’ cognitive responses, evidenced by their eye movements, showed that workers had significantly varied cognitive processes and abilities depending on the format of stimuli: Workers with lower hazard identification skills were more likely to miss hazards in a dynamic environment. This result suggests that the experimental setting should be carefully designed to determine construction workers’ natural cognitive process. 

Roofers spend considerable time in awkward postures due to steep-slope rooftops. The combination of these postures, the forces acting on them, and the time spent in such postures increases the chance of roofers developing musculoskeletal disorders (MSDs). Several studies have connected these awkward postures to potential risk factors for injuries and disorders; however, existing models are not appropriate in roof workplaces because they are designed to assess work-related risk factors for general tasks. This study examines the impacts of work-related factors, namely working posture and roof slope, on kinematics measurements of body segments in a laboratory setting. To achieve this objective, time-stamped motion data from inertial measurement unit (IMU) devices (i.e., accelerometer, gyroscope, and quaternion signals) were collected from a sample of six undergraduate students at George Mason University. Participants performed two common roofing activities, namely walking along the roof and squatting in different roof slopes (0°, 30°). Comparing IMU signals using statistical analysis demonstrated significant differences in body kinematics between roofing activities on the slope and level ground. Overall, sloped-surface activities on a 30° roof resulted in changes in about 26% of walking and 12% of squatting variables. Such information is useful for a logical understanding of roofing MSD development and may lead to better interventions and guidelines for reducing roofing injuries. 

Safety training has long been considered a promising method to enhance workers’ hazard identification skills within construction sites. To improve the effectiveness of safety training, such varied features as a training environment, individuals’ learning ability, and lesson personalization have been investigated. However, as records show workers still miss hazards even after receiving safety training, understanding the fundamental cognitive reasons for unrecognized hazards becomes a crucial step toward developing effective personalized safety training. This study used various 360° panoramas of construction scenarios to empirically examine 30 workers’ visual search strategies and assess workers’ hazard identification skills. Results suggest several cognitive limitations caused failures in hazard recognition, including attentional failure, inattentional blindness, and low perceived risk. Based on these findings, this study proposes a personalized safety training framework to address such cognitive limitations to improve occupational safety in the construction industry