Productivity Analysis of Precast Concrete Operations by Artificial Neural Networks

Productivity estimation is a crucial task for construction managers and general estimators to efficiently allocate the required resources in order to minimize the construction costs. There is a lack of research in terms of productivity modeling for precast erection process. Therefore, this study was designed to develop a model based on Artificial Neural Networks (ANN) to predict the installation times of the most commonly used precast elements namely: walls, columns, beams, and slabs. Installations of 220 precast elements were observed and significant factors influencing productivity were identified through stepwise Multiple Regression Analysis (MRA) to form the inputs of ANN model. Performance of the developed model on the test data showed its accuracy in predicting installation times of different precast components which confirmed the appropriateness of the model to be used by practitioners or construction management research scholars.