Topological indices encode molecular structure into numerical descriptors, enabling prediction of physicochemical properties without extensive experimentation. This study evaluates the implementation of graph-theoretical topological indices in combination with multi-criteria decision-making methods to compute physicochemical properties and ranking of eye drugs. A set of classical topological indices, consisting of N SK1, N SK2, M1, M2, N M1, N M2, N SK, N F , and N H, are computed using an edge-partition approach to describe the structural aspects of 12 selected drugs. Multiple linear regression models were developed to develop quantitative structure–property relationship models for important physicochemical properties, including molar volume, polarizability, molar weight, and molar refraction. The multiple linear regression models showed reliable predictive output, with statistical indicators and correlation coefficients higher than 0.89, confirming the value of the selected indices. The VIKOR technique enables the ranking of drugs based on their estimated physicochemical properties. The combined application of quantitative structure–property relationship and multi-criteria decision-making framework enhances robustness in discriminating structurally similar eye drugs. This combination offers a systematic framework for prioritizing compounds in pharmacological research and drug optimization. Such approach also supports public health by facilitating more efficient drug development and therapeutic decision-making for eye diseases.