The process of modeling elementary use cases

[fatass]

A definition:

An elementary use case is a use case which is executed under the
control of one primary actor, that starts and ends without any
intervening time delays.

Modelling elementary use cases - The steps:

1. Functional decomposition. Although a contentious technique to use
in the object oriented world, it serves well as a technique to get to
elementary use cases. By functional decomposition we mean breaking
down high level functions into it constituent lower level functions
until the elementary use case level is reached.

The higher level functions may be shown as high level use cases, but
it makes more sense to use package diagrams to show the decomposition.
At the elementary level we can use case diagrams to show the
elementary use cases and the associated actors.

2. Identify primary actors. Each elementary use case should have only
one primary actor. This will be the actor that initiates the use case
and controls the use case through the execution until it achieves the
goal of the use case.

3. Identify other actors/stakeholders. Other actors, or stakeholders,
are normally recipients of outputs of the use case interaction.

4. Documenting use cases. Simple documentation for the use cases may
include:

- Use case - A contract for the behaviour of the function
- Scope - The function under discussion
- Business rules - The rules that govern behaviour of the function
- Scenarios - The possible behaviour paths of the function
- Extensions - Descriptions of additional behaviour that enhances the
defined function

In order to ensure that modellers model at the same level of
abstraction and what should be modelled and documented, it should be
clearly defined in the UML standards of the enterprise.