This study sought to develop a typology model of temporary facility constraints that can be used practically in construction site layout planning (CSLP) automation models. A triangulated methodology (literature review, in-depth interviews, and actual case studies) was used to identify constraints of 11 temporary facilities that are considered mainly in CSLP and to classify them into six constraint types (i.e., dimensional, regional, relocation, non-overlap, inter-facility distance, and visibility constraints) and seven subtypes. In addition, this study proposed computational modeling methods that would allow a computer to judge whether or not the constraints are met by a created construction site layout. This study contributes to CSLP theory by providing a typology model of temporary facilities that can be used in the model-based verification of the created construction site layout and constraint processing condition in CSLP optimization models. This would also help provide efficient, safe, and eco-friendly construction site management, while refraining from experience- and intuition-centered CSLP practices.

The accident rate in the Korean construction industry is rapidly increasing each year. In particular, the mortality rate from human accidents in construction is very high compared to that of other industries. This indicates that the safety activities carried out at the national level have not been effective in reducing the accident rate. To resolve these problems, the risk factors for human accidents should not only be forecasted in advance, but should also be controlled, monitored, and managed from the commencement to the completion of construction projects. However, most studies on this topic have only listed the importance of risks based on the frequency of safety accidents rather than providing a quantitative analysis. Therefore, it is necessary to develop a model that can analyze and quantify the pattern of human accident risks. The purpose of this study is to develop a quantification model for human accidents in South Korean construction sites. To this end, insurance claim payout data were collected and analyzed to develop a risk quantification model using a Monte Carlo simulation. The results will be used as base data to develop a construction site safety management checklist and human accident prediction model.