Prediction of Concrete Properties Using Ensemble Machine Learning Methods

Prayogo, Doddy and SANTOSO, DAVID IMMANUEL and WIJAYA, DENNIS and GUNAWAN, TITANIA and WIDJAJA, JESSICA AGUSTINE (2020) Prediction of Concrete Properties Using Ensemble Machine Learning Methods. [UNSPECIFIED]

	PDF Download (936Kb)
	PDF Download (1587Kb)

Abstract

One of the most commonly used materials in civil engineering is concrete; not only is it cheap and strong, but it is also efficient and convenient. The efficiency of concrete is based on the easiness to place and to compact, which is usually known as workability. However, concrete strength and workability works in different ways; hence it is important to divide concrete into two groups: concrete with low workability and concrete with high workability, in order to achieve a more accurate prediction. Since there is a lot of variations of concrete mix designs, the relationship between each mixture is complex and, thus, requires advanced prediction methods in order to find the most accurate relationships between concrete mix proportion and its compression test result.�Recently, many studies have been conducted on applying multiple artificial intelligence (AI) methods in building different complex and challenging prediction models. Thus, this research employs ensemble machine learning methods to precisely forecast compression strength of concrete mix proportion. The accuracy of the proposed method was calculated using two performance measurements. Subsequently, the study has successfully built the prediction model that can accurately map the relationship between concrete mix proportion and compressive strength.

Actions (login required)