Sustainable production in cement via artificial intelligence based decision support system: Case study


Ates K. T., Şahin C., Kuvvetli Y., Kuren B. A., Uysal A.

CASE STUDIES IN CONSTRUCTION MATERIALS, vol.15, 2021 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 15
  • Publication Date: 2021
  • Doi Number: 10.1016/j.cscm.2021.e00628
  • Journal Name: CASE STUDIES IN CONSTRUCTION MATERIALS
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Compendex, INSPEC, Directory of Open Access Journals
  • Keywords: Decision support system, Ground granulated blast furnace slag, Prediction of activity index at 28 days, Sustainable CEM III production, BLAST-FURNACE SLAG, COMPRESSIVE STRENGTH, NEURAL-NETWORKS, CONCRETE, PREDICTION
  • Çukurova University Affiliated: Yes

Abstract

CEM III cement is the mixture of ground granulated blast furnace slag (GGBS) and Portland cement clinker. Portland cement clinker has been effectively substituted by GGBS in order to improve strength and performance concrete supplying both economic and environmental benefits such as resource and energy savings. The activity index of the GGBS is positively correlated with the 28-day compressive strengths of CEM III mix cements. However, it is necessary to wait for a test period of 28 days for the acquisition of the activity index value of granulated blast furnace slag. This situation negatively impacts the cost of production, product quality and the environmental impacts of the product. Therefore, it is required to have a decision support system that can predict the activity index without waiting for the test process of 28 days using some parameters obtained physical and chemical analysis at current. The paper aims to develop a decision support system for the prediction of activity index of granulated blast furnace slag at 28 days. Prediction abilities of artificial neural networks (ANN) and adaptive neuro-fuzzy inference systems (ANFIS) are investigated for the decision support system. According to the artificial neural network results, the mean absolute percentage error value for the whole data set has been calculated as 2.27 %. The comparisons show that ANN overcomes ANFIS approach. According to these results, the proposed decision support system has helped the determination of mixtures optimally and led to the decrease in CO2 emission of CEM III production process.