Designing a
Method Network (Hierarchy)
2.
Create a prototype for each process identified in the data
flow diagram in order to define input and output requirements as
well as to track method dependencies.
3.
Identify any standalone program units in the system, and
trace their relationships to other program units and data objects.
4.
Perform an initial mapping of method prototypes to data
elements.
There are two ways to proceed once the listed
steps have been performed. A top-down or a bottom-up strategy can be
implemented. The disadvantage of the latter is that it is difficult
to ascertain early in the effort which atomic-level program units
are needed to fulfill the project requirements. The top-down
approach is effective because only those processes required to meet
the business objectives will be targeted. As more facts surrounding
the relationship between the methods are uncovered, more of the
sub-methods will be moved to different parts of the network. As we
stated earlier, the effort starts with the data flow diagram or
functional model. The data dictionary is used in conjunction with
the data flow diagram to associate the atomic-level attributes
participating in the data process flow. At this stage, processes
have a much broader scope, and it is fully understood that further
decomposition will yield further sub-processes.
