BIM Reliability Prediction Method for Construction Schedule of Prefabricated Hollow Floor Based on Heterogeneous Data Fusion

In order to improve the prediction ability of the construction progress of a prefabricated hollow floor, a building information modeling (BIM) reliability prediction method for the construction progress of prefabricated hollow floor based on heterogeneous data fusion is proposed. Considering various uncertain factors in the actual construction process, the artificial neural network is used to simulate and calculate the construction project progress, and the mathematical model for the construction progress prediction of the prefabricated hollow floor is established. The time dimension is added to the three-dimensional (3D) model for construction simulation, and the key nodes in the construction process are found through the construction animation and model information; thus, the construction schedule is designed and adjusted. The heterogeneous data fusion method is used to realize the BIM information fusion and feature clustering analysis of the construction progress of the prefabricated hollow floor, extract the characteristic quantity of the association rules of the construction progress, realize the convergence control and self-adaptation learning training in the construction progress prediction process through the fuzzy neural network learning algorithm, and use the BIM4D platform to realize the visual simulation of the construction progress prediction and plan management. On this basis, the schedule shall be prepared. The experimental results show that this method has strong generalization learning ability and good visual prediction accuracy for the construction progress BIM prediction of prefabricated hollow floor, which effectively ensures the efficiency of project implementation and the quality of the project and also carries out corresponding cost control.