Development of Project Schedule Risk Analysis Based on Bagging Algorithm

Project progress risk analysis is critical to ensuring project delivery on time. The aim of this study is to improve the prediction accuracy of project schedule risk through Bagging algorithm and provide scientific decision support for project managers. The research uses Bagging and decision tree to model the project schedule related factors, and cross validation method is used to evaluate the performance of the model. The results show that Bagging is superior to single decision tree model with high accuracy (92.5%) and low mean square error (0.08) in risk prediction. It can be used as a tool for project managers to deal with progress risks.