Back to Search Start Over

Design and analysis of a high-rise building and its cost prediction using machine learning.

Authors :
Nanduri, Mohana Deepthi
Poojitha, V.
Sujatha, T.
Reddy, S. Rajashekar
Swaraj, J.
Source :
AIP Conference Proceedings; 2024, Vol. 3146 Issue 1, p1-7, 7p
Publication Year :
2024

Abstract

The main ideal of this document is to Analyze and Design of a Multi storey structure applying a Software usually known as STAAD Pro and its cost prediction using a AI technique called Machine Learning. In this we are designing a multi storey building of G+12 which involves load calculations, sesmic load conditions and analysing of building by STAAD pro and also analyzing the building erect for chancing the bending moments, shear forces, deflections and also underpinning details for the structural members of structure(similar as shafts, columns & crossbeams) for developing the provident and sustainable design. For this design process we used Indian standard code books of IS 456-2000, IS 875 (Part 1,2,3,5). AUTOCAD software is used for development of plans and the plan is imported to staad pro and designed. The total height of the building is 39m and the area of the building is around 8000sq.ft consisting of 12 floors. This paper also introduces machine learning (ML) technology to optimize design costs. In our project we used linear regression algorithm. Linear regression relationships are modelled using a linear prediction function that estimates unknown model parameters from data. Data is given in form of datasets that comprises of attributes such as essential building materials and their cost according to the area, so that the future cost is predicted accordingly. The result obtained in trained model and the manual estimation is accurate. In this project we are calculating and comparing the cost of high rise, mid rise and dwarf buildings through machine learning as well as through manual estimation of the building. [ABSTRACT FROM AUTHOR]

Details

Language :
English
ISSN :
0094243X
Volume :
3146
Issue :
1
Database :
Complementary Index
Journal :
AIP Conference Proceedings
Publication Type :
Conference
Accession number :
178559726
Full Text :
https://doi.org/10.1063/5.0224688