The Application of SOFM Fuzzy Neural Network in Project Cost Estimate

Applications of neural network were widely used in construct project cost estimate. Aim at handling weakness of poor convergence and insufficient forecast, an improved method based on SOFM (self-organizing feature map) was proposed to replace the fashionable T-S fuzzy neural network. The method illustrated how to apply SOFM algorithm to improve the fault such as poor convergence and insufficient forecast, with the considering of analysis in basic principle of fuzzy neural network. After optimizing of T-S fuzzy neural network model, construct project cost estimate model had been built up. Finally, the model was set up with the purpose of comparing generalization ability by 18 examples and 2 testing samples. Comparing the simulation, a positive result was found that SOFM fuzzy neural network had a better performance in reducing the forecast error and iterating times. Therefore, this model is fit for handling construct project cost estimate.