Thermal conductivity is one of the most important properties of materials especially liquids. Thermal conductivity is highly interesting since it is the main parameter required for most heat-transfer calculations or design of industrial equipments such as heat exchangers. Unfortunately, thermal conductivity is extremely hard to be experimentally measured due to some operational issues which shifted the researchers’ interest toward proposing a correlation to predict this property. In the light of this limitation, in the current study, a four-parameter correlation based on critical temperature ([Formula: see text], critical pressure ([Formula: see text] and acentric factor ([Formula: see text] of substances is proposed to predict thermal conductivity of liquids. In this way, 956 experimental data points collected from previously published literatures were divided into two different subsets, namely training and testing subsets, in the first stage and then used to find the optimum values of fitting parameters of the proposed correlation. Based on the obtained results of error analysis, it can be concluded that the proposed correlation has capability of both extrapolating and correlating the thermal conductivity of liquids.