Prediction Model of Construction Accidents During the Execution of Structures Using Decision Tree Technique

Abstract

Construction is recognized as a dangerous industry. For this reason, in the Literature can be found a large number of studies with the aim of identifying the causes in order to devise mitigation strategies. In recent decades, the emergence of Information and Communication Technologies is providing a new research framework with this objective. This framework includes different techniques that allow, among other things, the analysis of large amounts of data efficiently. In this work, occupational accidents in Spain from 2003 to 2015 are explored. The objective of the present study is to develop a predictive model which not only could predict the occupational incidents but also provide a decision tree view in order to analyze the model. To achieve this objective, the KNIME Analytics Platform is used. The obtained results are good since a 99% of accuracy is obtained.