Analysis of Labor Productivity of Brick Masonry Work in Building Construction in Kathmandu Valley

Construction labor productivity is the most determinant of success of any construction project. Labor is considered as more variable and unpredictable cost component for the successful accomplishment of construction projects. The main aim of this research is to develop an artificial neural network (ANN) model to predict the production rate for brick masonry work by assessing the various factor affecting labor productivity. Out of forty-four factors selected from a literature review, the top thirteen factors were selected for model development after the questionnaire survey and ranking them based on Relative Importance Index (RII). The model was developed in Neurosolution version 7.1.1.1 using the various input data set collected from active construction site of brick masonry. 65% of data set were used for training, 20 % of data set were used for cross-validation and remaining 15 % of data set were used for testing. The error between actual productivity and estimated productivity was computed using Mean Square Error (MSE) which was 0.019 which verified that the estimated production rate was within an acceptable range. After the successful testing of model, a sensitivity analysis was performed to analyze the order of most influencing factors affecting labor productivity. The developed ANN model can be used for estimating the labor productivity of brick masonry work for any building construction project by incorporating the influence of selected parameters or factors.