The purpose of this study is to provide a method for classifying essential oils based on the physicochemical properties and the composition ratios of essential oils components as a basic study for selecting essential oils that are suitable for individual body constitution and symptoms of disease. Twenty nine essential oils, of the main essential oils used in aromatherapy, with complete data values of 4 physicochemical properties (molecular weight, solubility, flash point, and relative density) were selected to conduct a cluster analysis. As a result, 29 essential oils were classified into 4 clusters. Cluster 1 was found to have the highest molecular weight, solubility and flash point and the second highest density compared to other clusters. Cluster 2 was found to be higher than average in terms of molecular weight, solubility, flash point with the highest density. Cluster 3 showed higher molecular weight and flash point than the average, whereas solubility and density are lower than the average. Cluster 4 showed the lowest molecular weight, solubility, flash point, and density. It was found that the results of the classification based on the physicochemical properties of chemical constituents were not fully consistent with the results of the existing classification methods of essential oil. Through the result of this study, further study regarding the classification is required to be conducted in the future and will become a basic research that could help to select individual customized essential oils with the new perspective of essential oil classification using the physicochemical properties. Keywords: Aromatherapy; Cluster analysis; Constituent; Essential oil; Physicochemical properties