MODELING AND OPTIMIZATION OF STANDARD CONCRETE CONTAINING GRANULE BLAST FURNACE SLAG: A GENE EXPRESSION MODELING BASED MULTI-RESPONSE WEIGHTED NON-LINEAR PROGRAMMING APPLICATION

In this study, Gene expression models have been obtained for prediction of slump flow, 3rd day compressive strength, 7th day compressive strength and 28th day compressive strength of normal weight concrete containing granule blast furnace slag using GeneXproTools software package version 5.0 with data collected from concrete quality control laboratory. Optimal mixture proportions of normal weight concrete containing granule blast furnace slag were determined by multi-response weighted non-linear programming using using non-linear gene expression models via Matlab version 2015a. It is concluded that proposed gene expression models based multi-response weighted non-linear programming methodology is the first application that is effective in order to determine optimal mixture proportions when compared to other methods such as fuzzy modeling and artificial neural network. In addition, this methodology provides the possibility of optimal use of resources such as raw materials for laboratory personnel without the need for additional experiments causing a loss of time.