Journal of Artificial Intelligence 1994-5450 2077-2173 Asian Network for Scientific Information 10.3923/jai.2011.63.75 Arafa Mohammed Alqedra Mamoun 1 2011 4 1 Early stage cost estimate plays a significant role in any initial construction project decisions, despite the project scope has not yet been finalized and still very limited information regarding the detailed design is available during these early stages. This study aimed at developing an efficient model to estimate the cost of building construction projects at early stages using artificial neural networks. A database of 71 building projects is collected from the construction industry of the Gaza Strip. Several significant parameters were identified for the structural skeleton cost of the project and yet can be obtained from available engineering drawings and data at the pre-design stage of the project. The input layer of the Artificial Neural Networks (ANN) model comprised seven parameters, namely: ground floor area, typical floor area, number of storeys, number of columns, type of footing, number of elevators and number of rooms. The developed ANN model had one hidden layer with seven neurons. One neuron representing the early cost estimate of buildings formed the output layer of the ANN model. The results obtained from the trained models indicated that neural networks reasonably succeeded in predicting the early stage cost estimation of buildings using basic information of the projects and without the need for a more detailed design. The performed sensitivity analysis showed that the ground floor area, number of storeys, type of foundation and number of elevators in the buildings are the most effective parameters influencing the early estimates of building cost.

]]> Adeli, H. and M. Wu, 1998 J. Construction Eng. Manage., 124 18 24 Alqedra, M.A. and A.F. Ashour, 2005 Comput. Struct., 83 2495 2502 Al-Tabtabai, H., P.A. Alex and M. Tantash, 1999 Construct. Eng., 41 19 24 Ashour, A.F. and M.A. Alqedra, 2005 Adv. Eng. Software, 36 87 97 Emsley, M.W., D.J. Lowe, A.R. Duff, A. Harding and A. Hickson, 2002 Construct. Manage. Econo., 20 465 472 Gunaydin, M.H. and S.Z. Dogan, 2004 Int. J. Project Manage., 22 595 602 Hagan, M.T., H.B. Demuth and M.H. Beale, 1996 Neural Network Design. 1st Edn., Harding, A., D. Lowe, M. Emsley, A. Hickson and R. Duff, 1999 Proceedings of the Construction and Building Research (COBRA), 1999 pp: 161 168 Kim, G.H., D.S. Seo and K.I. Kang, 2005 J. Comput. Civil Eng., 19 208 211 Orlando, D., R. Nibaldo and A.C. Luiz, 2009 Expert Syst. Appl., 36 7435 7440 Pettang, C., L. Mbumbia and A. Foudjet, 1997 Construct. Build. Mater., 11 47 55 Rafiq, M.Y., G. Bugmann and D.J. Esterbrook, 2001 Comput. Struct., 79 1541 1552 MATLAB, 2009 MATLAB the Language of Technical Computing. Version 2009B. Min-Yuan C., T. Hsing-Chih and H. Wen-Shan, 2009 Automation Construction, 18 164 172 Cheng, M.Y., H.C. Tsai and E. Sudjono, 2010 Expert Syst. Appl., 37 4224 4231 Ozgan, E. and R. Demirci, 2008 J. Applied Sci., 8 146 151 Setyawati, B.R., S. Sahirman and R.C. Creese, 2002 Neural networks for cost estimation. 2002 Shi, J.J., 2002 J. Comput. Civil Eng., 16 152 155 Solaimani, K., 2009 Asian J. Applied Sci., 2 486 498 Sonmez, R., 2004 Can. J. Civil Eng., 31 77 683 Wilmot, C.G. and B. Mei, 2005 J. Construction Eng. Manage., 131 765 771 Yang, K.H., A.F. Ashour, J.K. Song and E.T. Lee, 2008 Struct. Build., 161 29 39 Zhang, Y.F. and J.Y.H. Fuh, 1998 Comput. Ind. Eng., 34 433 450