Skip to main content
placeholder image

Application of PSO-based neural network in quality assessment of construction project

Conference Paper


Abstract


  • Construction project quality management, the basis of construction management, is crucial for construction firms to survive and grow in the industry. This paper presents the adoption of a particle swarm optimization (PSO) model to train perceptrons in assessment and predicting the quality of construction projects in China. Artificial Neural Network (ANN) has preeminent learning ability, but often exhibit inconsistent and unpredictable performance for noisy data. The Particle Swarm Optimization (PSO) technique is used to train the multilayered feed forward neural networks to discriminate the different operating conditions. Comparing with backpropagation ANN and ANN based on genetic algorithms, the simulated results of quality assessement of construction projects show that training the neural network by PSO technique gives more accurate results (in terms of sum square error) and also faster (in terms of number of iterations and simulation time) than BPN and GA-based ANN. © 2008 IEEE.

Publication Date


  • 2008

Citation


  • Shi, H., & Li, W. (2008). Application of PSO-based neural network in quality assessment of construction project. In Proceedings - 2008 International Conference on MultiMedia and Information Technology, MMIT 2008 (pp. 54-57). doi:10.1109/MMIT.2008.66

Scopus Eid


  • 2-s2.0-70349593769

Start Page


  • 54

End Page


  • 57

Abstract


  • Construction project quality management, the basis of construction management, is crucial for construction firms to survive and grow in the industry. This paper presents the adoption of a particle swarm optimization (PSO) model to train perceptrons in assessment and predicting the quality of construction projects in China. Artificial Neural Network (ANN) has preeminent learning ability, but often exhibit inconsistent and unpredictable performance for noisy data. The Particle Swarm Optimization (PSO) technique is used to train the multilayered feed forward neural networks to discriminate the different operating conditions. Comparing with backpropagation ANN and ANN based on genetic algorithms, the simulated results of quality assessement of construction projects show that training the neural network by PSO technique gives more accurate results (in terms of sum square error) and also faster (in terms of number of iterations and simulation time) than BPN and GA-based ANN. © 2008 IEEE.

Publication Date


  • 2008

Citation


  • Shi, H., & Li, W. (2008). Application of PSO-based neural network in quality assessment of construction project. In Proceedings - 2008 International Conference on MultiMedia and Information Technology, MMIT 2008 (pp. 54-57). doi:10.1109/MMIT.2008.66

Scopus Eid


  • 2-s2.0-70349593769

Start Page


  • 54

End Page


  • 57