This paper proposes dimensionless correlations predicting properties of the frost layer formed on a cold flat surface. Experiments are carried out to obtain the correlations with various environmental parameters such as air temperature, air velocity, absolute humidity, and cooling plate temperature. As a result, the frost properties (frost layer thickness, density, surface temperature, thermal conductivity) are correlated as a function of Reynolds number, Fourier number, absolute humidity and non-dimensional temperature by using a dimensional analysis. The correlations agree well with the previous and our experimental data within a maximum error of 10%, and are used to predict the frost properties in the following ranges: Reynolds number of 20216 to 53763, Fourier number of 0.1962 to 2.5128, absolute humidity of 3.22 to 8.47, and non-dimensional temperature of 0.125 to 0.5.