Information systems have grown consistently larger and their functionality has become increasingly complex. In the analysis of a large, complex information system, the system functionality should be decomposed into sub-functions for the effective division of analysis, design and implementation from the early analysis phase. However, the object-oriented analysis as well as the structured analysis don’t successfully provide an obvious solution for effective functional decomposition of a large complex information system. For solving this problem which both the object-oriented analysis and structured analysis have in the aspect of functional decomposition of information systems, the hybrid methods, like FCD, have gotten a reasoning power. Those hybrid methods integrate structured analysis wit object-oriented approach. In this research, the theoretical mechanism of FCD, the leading hybrid method in functional decomposition methodology using scenario analysis is introduced. The original FCD method was improved in two aspects, measurement specification of scenario refinement and quantitative measurement of coupling level for class grouping. Using improved new model, executed functional decomposition of an online shopping mall system.
