Estimation of recycling capacity using ANN and SVM

Građevinar

article

Aleksandra Vujkov¹  Dragana Bibić¹  Igor Peško¹  Vladimir Mučenski¹  Jasmina Dražić¹  Milan Trivunić¹ 

[1] University of Novi Sad Faculty of Technical Sciences Department of Civil Engineering and Geodesy

关键词: buildings;    recycling;    concrete;    reinforcement;    artificial neural networks;    support vector machine method;   

DOI  :  10.14256/JCE.2273.2017

学科分类：社会科学、人文和艺术（综合）

来源: Hrvatsko Drustvo Gradevinskih Inzenjera

【 摘 要 】

This paper presents estimation of the quantity of concrete and reinforcement that can be recycled for residential buildings constructed as skeleton structures. Models based on artificial intelligence, involving the use of Artificial Neural Networks (ANNs) and the Support Vector Machines (SVM) methods, were formed in order to estimate quantities of these materials. The results show that the application of ANNs and SVM methods is a good solution for the estimation of recycling capacity. The mean absolute percentage error (MAPE) for the selected ANNs for predicting quantity of concrete and reinforcement is 8.74 % and 12.58 %, respectively.

【 授权许可】

CC BY