Improved BP Neural Network for Construction Schedule Prediction Under Lean Construction System.

To address the problem of low accuracy of construction schedule prediction in location-based management systems (LBMS), this paper adopts Harmony Search Algorithm (HS) to optimize the weights and thresholds of BP neural networks. By integrating practical construction information, an LBMS framework based on HS-BP neural networks is constructed to accurately predict the construction schedule. Finally, the prediction accuracy of HS-BP neural network is verified by taking the construction schedule of roof slab formwork of Guangzhou Guanglianda South China Headquarters Base Project as an example. The flow line of the actual and predicted schedules of the beam slab are also drawn to illustrate the reliability of the LBMS framework based on HS-BP neural network. [ABSTRACT FROM AUTHOR]