%P 713-726
%T Construction labor production rates modeling using artificial neural network
%V 16
%A S. Muqeem
%A A. Idrus
%A M.F. Khamidi
%A J. Bin Ahmad
%A S. Bin Zakaria
%O cited By 13
%L scholars2071
%J Electronic Journal of Information Technology in Construction
%D 2011
%K Artificial Neural Network; Average values; Concrete building structures; Construction productivity; Formwork; High rise; Influencing factor; Labor productivity; Learning capabilities; Modeling technique; Neural network prediction model; Prediction model; Production rates; Site conditions; Work sampling, Concrete construction; Database systems; Engineering research; Forecasting; Mathematical models; Mean square error; Pattern recognition; Productivity; Sampling, Neural networks
%X Construction productivity is constantly declining over a decade due to the lack of standard productivity database system and the ignorance of impact of various factors influencing labor productivity. Prediction models developed earlier usually neglect the influencing factors which are subjective in nature such as weather, site conditions etc. Many modeling techniques have been developed for predicting production rates for labor that incorporate the influence of various factors but artificial neural network (ANN) has been found to have strong pattern recognition and learning capabilities to get reliable results. Therefore the objective of this research is to develop a neural network prediction model for predicting labor production rates that takes into account the factors which are in qualitative form. The objectives of the research have been achieved by collecting production rates data for formwork of beams from different high rise concrete building structures by direct observation. Reliable values of production rates have been successfully predicted by ANN. The average value of 1.45xE-04 has been obtained for Mean Square Error (MSE) after testing the network. These results indicate that the ANN has predicted production rates values for beam formwork successfully with least range of errors. © 2011 The authors.