DATA SCIENCE HELPS TO SOLVE PROBLEMS OF MIX OPTIMIZATION AND QUALITY CONTROL
CONCRETE
UDC 691.3: 519.6
[:ru]Резаев Р. О.[:en]Rezaev, R. O.[:] [:ru]канд. физ.-мат. наук, доцент, Томский политехнический университет, г. Томск, Россия; научный сотрудник, Leibniz Institute for Solid State and Materials Research, Dresden, Germany[:en]Ph.D., associate professor, Tomsk polytechnic university, Tomsk, Russia; researcher in Leibniz Institute for Solid State and Materials Research, Dresden, Germany[:]
[:ru]Дмитриев А. А.[:en]Dmitriev A. A.[:] [:ru]главный технолог, «Проектирование материалов», Москва, Россия[:en]Chief Technologist, “Materials Design”, Moscow, Russia[:]
Alitinform №3 (68) 2022 г. 28-40 p.
Abstract
The complicated nature of concrete due to the complexity and variety of the materials systems involved, makes it generally impossible to design concrete mixes with desired properties relying on purely theoretical background and first-principle methods. Within a more utilitarian approach, one does not try to completely understand the causal physics and chemistry behind the processes having an impact on the target property of the design, but takes advantage of gathered statistics on that property. Within this approach, industrial statistics methods and data-science frameworks play a crucial role. The main part of this paper is dedicated to the approaches to tackling the concrete mix design problem. Special attention will be given to the Bayesian statistics approach, which may have a wide range of applications in the industry. In the last section, we outline a particular use of the Bayesian methods for the quality control problem.
Key words:
concrete mix optimization, ready mix concrete, quality control
Read more in the magazine
To do this you need to subscribe to one of the options and can read all the articles without limitation and in any format that is convenient for you.
Subscribe
Spelling error report
The following text will be sent to our editors: