This study investigates how variations in constituent materials affect the fresh properties of cement-based pastes using a statistically driven experimental approach. A Central Composite Design (CCD) was implemented to examine the influence of three key input parameters: water-to-cement ratio (w/c), superplasticizer-to-powder ratio (Sp/p), and water-to-powder ratio (w/p). Fifteen mix compositions were produced and tested using the mini-slump test and Marsh funnel flow time, both immediately after mixing and after 60 minutes. Response Surface Methodology (RSM) was applied to develop predictive models for each property. The results showed that the water-to-powder ratio was the most influential factor on workability, followed by the superplasticizer-to-powder ratio. The statistical models successfully captured main, interaction, and quadratic effects, enabling accurate prediction of flow and time measurements. These models were further used to optimize mix compositions according to targeted fresh-state performance. Compared with conventional one-variable-at-a-time approaches, the CCD method substantially reduces the number of tests required while providing deeper analytical insights. The proposed methodology improves the understanding of complex interactions among mix parameters and supports the efficient design of cement-based materials for performance-critical applications.
Lino Maia (Thu,) studied this question.