Abstract: Cost estimating is one of the most important and challenging activities during project planning, which occurs at the early stages of a project life where limited information is available and many unknown factors affecting the project costs. ANN is a new advent that is used in cost estimation, which is able to lucubrate from experience and examples and deal with non-linear problems. It can perform tasks involving deficient data sets, fuzzy or insufficient information and for highly complex problems. The objective of this study is to review cost estimation models which used Artificial Neural Network (ANN) tool and to suggest the most effectual algorithm for cost prediction and the factors predominantly affecting the total construction costs of building projects. To build CEM, the most effective factors affecting cost in construction projects were identified based on a comprehensive survey among a collected sample of construction relevant model studies. The developed neural network model examines the data set into distinct cases classified on the basis of hidden layers. Each of them containing the independent input neurons, hidden layers and a dependent output neuron. The results of the trained models indicated that neural network reasonably succeeded in estimating the Total construction cost of building projects at the planning stage itself. The average error of test dataset for the adapted model was largely acceptable and can perform as a good indicator regarding the ability of the proposed model to predict the total construction cost of any future construction project at an appreciated degree of accuracy. This paper gives a clear review of implementing the ANN tool in prediction of total cost of building construction projects and the relevant factors affecting it.

Keywords: Artificial Neural Network (ANN), Cost Estimation Model (CEM), effective algorithm, prediction.