Taguchi-based interval multi-criteria decision-making method for optimisation of alkali activators

Abstract

The main purpose of this study is to assess the optimal condition with multiple responses using the Taguchi method in the presence of conformity assessment. For this purpose, all possible arrays, including orthogonal and non-orthogonal arrays, were included in the analysis. The experimental data in the form of interval numbers were directly derived from experimental results for orthogonal arrays, and were computed using the Taguchi prediction method for non-orthogonal arrays. A new pattern of conformity assessment was introduced that eliminates arrays that do not meet the criteria or modifies the interval number of non-orthogonal arrays that meet a part of the criteria. Finally, each accepted array was specified as an alternative, and interval multi-criteria decision-making methods were finally deployed to rank the alternatives. A case study was carried out to find the optimal mixture of sodium silicate activators with strict criteria of ASTM C150 based on the proposed method. To this end, three factors, including the weight percentage of sodium oxide (Na2O) in slag, silicate modulus (SiO2/Na2O), and water to slag (W/S) ratio, each at three levels, were investigated by examining the compressive strength and drying shrinkage of the mortar, along with the setting time and mini-slump cone tests on the fresh paste. Results revealed the effectiveness of the proposed approach and indicated that the Na2O, SiO2/Na2O, and W/S values of 6%, 0.8, and 0.42, respectively, were a reasonable choice for the activator.