Estimasi Dimensi Optimum dan Rasio Tulangan Gedung Hotel Yogyakarta dengan Artificial Neural Network

Abstract

Cost estimation at the conceptual design stage of a building is necessary because it is considered as a fundamental input in further decision making. This paper presented a neural network method in approximating the optimum dimension and the minimum reinforcement ratio of beams and columns at the conceptual design stage of hotel buildings in Yogyakarta. A group of 27 building variations were prepared as the training data for the set up neural network model. Fourteen empirical formulas were obtained which could be used to estimate the optimum dimension and minimum reinforcement ratio of beams and columns with 5 parameters which consist of soil sites class, beam spans, number of floors, concrete strengths and diameters of the reinforcing bar.