Dependency Model
As defined by Seidel et al. we model a goal as a sequence of objectives. The Dependency Model allows ordering the objectives which are represented by the circles in the dependency graph.
Description: The designer should use state transition diagrams to model dependency models. Supported notation elements are: Objectives with labels and transitions without labels in one direction between them. The existence of a designated start state is also essential.
Description: For representing a goal, the designer should use a connected and directed graph of objectives. Therefore every objective should be directly or indirectly connected to every other one.
Example: To give an impression about how a dependency model can look like, please have a look at the figure shown below.
Description: Every goal has a so called start state that is neither renamable nor deletable. It describes the initial situation and in order to fulfill guideline DEP2, especially the start state must also be included into the dependency graph.
Description: The designer should make sure to only create acyclic graphs of objectives to obtain clarity of their processing order.
Example: When there is a cycle in the dependency model as shown in the figure below it is not clear which objective is the first one to be fulfilled. They're not only missing a clear semantic but also an incoming transition from the start state as well as an outgoing transition to the final state. In this case we would therefore also violate both guidelines DEP3 and DEP4.
Description: The designer should use meaningful names for each objective. Objective names should also be unique within one goal.
Description: The designer should enrich objectives by deadlines. These deadlines define a date relative to the start of the project when the objective shall be fulfilled.