Identifying and quantifying construction safety risks at the attribute level

The number of injuries and fatalities is disproportionally high when compared with other industries. In addition to physical pain and emotional suffering experienced by the victims and their families, these incidents have staggering societal costs. Therefore, investing in construction safety and developing innovations that improve safety is critical. The dissertation includes five manuscripts. The first explores the diffusion patterns of traditional injury prevention practices using common innovation diffusion models. The implications of the findings are that the construction industry has now reached saturation with respect to traditional injury prevention strategies and new safety innovations are needed. One of the most recent advancement in the preconstruction safety management strategies, that is proved to be highly effective, is to integrate safety risk data in to the schedule of project. Therefore, the second and third papers identify safety risks of common highway construction work tasks and their temporal and spatial interactions using the Delphi method and integrate them into a decision support system to produce predictive plots of safety risk over time based on the temporal and spatial interactions among concurrent activities. While, the results indicate that integrating safety risk data with schedule of project is highly effective, using the current methods to quantify safety risks for every individual task that can be experienced is infeasible with current risk modeling and data collection approaches. To address this limitation, the forth paper presents an attribute-based risk identification and analysis method that helps safety managers to identify and model the safety risk independently of specific activities or trades. The fundamental attributes that cause accidents are identified and their associated risks quantified by conducting reliable content analysis on 1771 accident reports from the National databases. The last paper uses the attribute-based risk management concept and proposes several safety predictive models to determine the outcome of possible injuries in early phases of a project. This research yield robust data and mathematical forecasting models that can be to objectively, accurately, and reliably predict hazardous conditions based on the identifiable attributes that characterize the workplace. It is expected that the findings of this research will transform the current risk analysis techniques and the created database have the potential to be applied to information models and emerging construction technologies.