Applying Clustering Algorithms to Explore the Circumstances of Fatal Falls in Construction Industry

摘要

This research adopted constrained k-prototypes algorithm to analyze the database that contains 294 cases of fatal falls between the years 2003 and 2010 within the Taiwanese construction industry. The objective of this study is to explore the circumstances of the fatal falls. The results of the analysis indicate that the primary circumstances under which the accidents occurred include the professional category of other specialized construction businesses; hazardous mediums such as scaffolding, support frame, stairs, stairways, and the openings, etc.; the time period between 9:00 am and 11:00 am; the falling height between 0 and 10meters. Meanwhile, not adopting safe construction methods is a major indirect factor that contributes to the accidents. The fundamental factors contributing to the accidents include not clearly informing the subcontractors of the conditions of the work environment, risk factors, labor health and safety regulations as well as the necessary measures to ensure the health and safety prior to the project commences.