Quantitative structure-solubility relationships (QSSR) are considered as a type of Quantitative structure-property relationship (QSPR) study in which aqueous solubility of chemicals are related to chemical structure. In the present work, multiple linear regression (MLR) and artificial neural network (ANN) techniques were used for QSSR studies of the water solubility of 68 phenols (phenol and its derivatives) based on molecular descriptors calculated from the optimized 3D structures. By applying missing value, zero and multicollinearity tests with a cutoff value of 0.95, and a genetic algorithm (GA), the descriptors that resulted in the best fitted models were selected. After descriptor selection, multiple linear regression (MLR) was used to construct a linear QSSR model. The R 2 = 91.0 %, LOO 2 Q = 89.33 %, s = 0.340 values of the model developed by MLR showed a good predictive capability for log S values of phenol and its derivatives. The results of MLR model were compared with those of the ANN model. the comparison showed that the R 2 = 94.99 %, s = 0.245 of ANN were higher and lower, respectively, which illustrated an ANN presents an excellent alternative to develop a QSSR model for the log S values of phenols to MLR. age to the environment, plants, animals and human health. The compounds penetrate ecosystems as the result of drainage of municipal or industrial sewage to surface water. 4 Therefore, it is vital to protect the environment and prevent their behavior by studying their physicochemical properties.Aqueous solubility is the concentration of a chemical in the aqueous phase, when the solution is in equilibrium with the pure compound in its usual phase (gas, liquid or solid) under standard conditions of temperature and pressure. 5 Aqueous solubility is one of the major physiochemical properties to be optimized in pharmaceutical and environmental studies; it is related to absorption and distribution in ADME-Tox (absorption, distribution, metabolism, excretion-toxicity).Experimental determination of compound solubility is not easily managed, or even possible, when working with large chemical libraries. 6 Alternately, the quantitative structure-property relationship (QSPR) provides a promising method for the prediction of solubility using descriptors derived solely from the molecular structure to fit experimental data. The QSPR approach attempts to establish simple mathematical relationships to describe the correlation of a given property to molecular structures for a set of compounds. 7 The advantage of this method lies in the fact that it requires only knowledge of the chemical structure and is not dependent on any experimental property. Moreover, it could be used for the prediction of the properties of new compounds.Recently, several works reported QSPR studies on the solubility and other properties. Warne et al. 8 reported the utility of thirty-nine molecular descriptors and physicochemical properties to model the solubility (S) and octanol-water partition coefficient (K ow ) of thirty-one lipophilic or...