| Abstract: | This dissertation documents and presents a series of studies conducted by the author at Texas A&M University with the overall goal of understanding the human-technology frontier from the perspective of the future workforce. Although construction is one of the oldest industries globally, the digital innovation and application of artificial intelligence (AI) in construction are still insignificant. Many trades within this industry are only beginning to realize the actual value of AI and automation in their daily workflow processes to improve safety, performance, and productivity. However, human workers will still collaborate with AI in many construction tasks for the foreseeable future. In this dissertation, through a systematic literature review of how AI impacts human workers' performance, behavior, and experience in the construction field, this review identifies the common human factors affected by the introduction of AI in construction. One of the rapidly emerging technologies in construction is the unmanned aerial vehicle (UAV), commonly known as the drone. Drones are increasingly used for site surveying, safety inspection, and progress monitoring and verification. However, human operators still play an essential role in the safe operation of drones. Therefore, this dissertation also presents a series of studies conducted with the overall goal of understanding and quantifying the drone operator's physiological state as a predictor of drone accidents. In particular, the reliability of drone operators' physiological indices and self-assessments to predict performance, mental workload (MWL), and stress in immersive virtual reality (VR) training and outdoor deployment is examined. Deep learning models are trained and tested with time-series physiological signals to predict potential drone accident events in (near-) real-time. As the construction workforce is aging, the industry is struggling to attract young job seekers and faces a lack of an energetic and new skilled workforce. This gap in the skilled workforce has also opened a new opportunity for nontraditional workers to enter the construction industry. The increasing influx of nontraditional employees and part-time workers, coupled with new expectations about alternative work arrangements (e.g., remote work), necessitates the identification of career-related strengths and weaknesses to match prospective employees with new job opportunities. Therefore, the focus of the last chapter in this dissertation shifts to investigating key challenges and opportunities associated with a career change for a nontraditional workforce. Keywords: Physiological signal, Machine learning (ML), Deep learning (DL), Accident prediction, Construction safety, Unmanned aerial vehicle (UAV), Virtual reality (VR), Wearable technology, Human performance, Mental workload, Veterans' strengths, Veterans' weaknesses, Veterans' support system, Thematic analysis The electronic version of this dissertation is accessible from https://hdl.handle.net/1969.1/198781 |
|---|