Applying Principal Component Analysis and Unascertained Method for the Analysis of Construction Accident Risk

In order to evaluate the construction accident risk, a novel method integrating principal component analysis (PCA) and the unascertained number algorithm with the synthesis weights is presented. According to humanmachine engineering, the multi-index and hierarchical system of accident risk in construction projects was built. PCA is applied to find relevant factors of the accident risk, which are classified into 6 categories used as inputs of unascertained assessment model. To improve the accuracy, quantitative change and qualitative change all need be reflected in the assessment results of accident risk in construction projects, the method determining the synthesis weights of indexes including both subjective favoritism and objective information is presented so as to make full use of redundant information, therefore, qualitative knowledge and quantitative data could be integrated. Finally, an improved unascertained assessment approach model was constructed to assess accident risk in construction projects by combining qualitative and quantitative information over conventional or other subjective methods and was verified by a case study.