The objectives of this study was to discover knowledge in predicting lengths of stay of Cesarean Section by patients characteristics and treatment method using data mining technique and to suggest the approach to the development of critical pathway. The findings suggest that data mining technique from the large pool of accumulated patients data can be utilized to systematize newly observed correlations, patterns and trends and to develop critical pathway for the treatment and management. The results of the study can be contributed to aid developing the clinical pathway for cesarean section suitable to Korean patients. And the application of the developed critical pathway in clinical practice will produce the actual effect and value.