Estimation of rebar weight in residential buildings of concrete intermediate moment frame using Artificial Neural Network

Estimating the cost of building construction, especially in the early stages of studies, is a topic of interest and importance for employers and investors. The high inflation of the country's economy in recent years and the sharp fluctuations in the price of construction materials have doubled the importance of estimating the cost of building construction, even in small urban projects. However, until the design of the building is completed and the final plans are not prepared in full detail, it is not possible to measure the project and estimate the construction costs with acceptable accuracy. On the other hand, the proposed approximate methods for initial estimates are associated with many differences from reality that are not consistent with the existential philosophy of project economic estimation. Therefore, in this study, using training capability of Artificial Intelligence and Artificial Neural Network algorithms, the weight of rebar used in concrete buildings of concrete moment frame, without complete design and preparation of executive plans is calculated and the results are compared with real values. The proposed artificial neural network model is a multilayer feeder type feeder with post-diffusion learning algorithm and is based on the parameters of number of openings in longitudinal and transverse direction, floor height, number of floors, number of columns per floor, floor area and seismic base shear. The weight of rebar used in the studied buildings has been estimated. The results indicate that the proposed neural network model can estimate the weight of rebar used in regular buildings with 95% accuracy and in irregular buildings with 80% accuracy.