Heat transfer processes account for a signifcant portion of the world’s total energy consumption, making it crucial to enhance their effciency. Nanofluids have garnered considerable attention in this domain as a means to achieve this objective. This study focuses on investigating the impact of key variables, including temperature, concentration, and specifc surface area of graphene nanoplatelets (GNPs) dispersed in distilled water, on the thermal conductivities of these nanofluids. The quantitative structure–property relationship (QSPR) methodology is employed for this analysis. The experimental conditions in preparing GNP-based nanofluid are encoded using the quasi-SMILES notation. The calculated optimal descriptors were acquired using the Monte Carlo method in CORAL software. The results of the best model indicated that the R2 m values for all splits are more than 0.5 and ΔR2 m less than 0.2. The predictive prowess and reliability of the models were substantiated by the statistical characteristics. For the best model (split 2), these values are R2 m= 0.9290 and ΔR2 m= 0.0301, which reveals the quality of the model. The created models had a clearly defned application domain, and correlation weights were calculated for all splits. The analysis of calculated correlation weights for the developed model are used to identify the main effects of experimental variables. The obtained results indicated that high concentrations, high temperatures, and high surface areas positively influenced the thermal conductivity ratio of the investigated carbon-based nanofluids. Conversely, low temperatures and low surface areas had a negative impact on the target property.
