Using artificial intelligent techniques in precast construction project progress/ schedule estimation

Abstract

Determining construction schedule of Prefabricated construction is an important issue for investors and Prefabricated Contractors. In terms of Construction characteristics, there are many factors that affect the progress of assembly, so Classification and Regression Trees (CARTs), ANN (Artificial Neural Networks), SVM (Support Vector Machine) and Ensemble are used to solve this problem. In this study, four models are constructed to forecast the progress of the assembly. Fifty data Constructions have been collected. The Cross Validation method is applied to predict the result. With four models built, the SVM model gives the best results with low error and convergence capability to determine the progress of assembly.