The system engineering based consumer’s behavior analysis methods are developed. The methods consist of the three models described in different viewpoints: process, status, and information structure. Consumer’s behaviors for purchasing soft drinks are analyzed with these methods in a hierarchical process as “totality to individual”. The behavior as a group is analyzed with aggregation data by statistical methods. Then, the discovered knowledge is explained by analyzing the behavior as an individual with the three above-mentioned models. Thus, spiral dynamics of consumer’s behavior generated by repetitive fluctuations of the concern is detected.