Process Modeling - Event-oriented model (Event-Driven)

1 - About

An event-oriented model the system from an event point of view (not from an entity point of view). Ie when a event occurs, which transformation happens to the entities.

Instead of focusing on current state, you focus on the event (changes) that have occurred over time.

The event (that starts the flow/chain) must get the system moving by changing its state (ie which event will get the system moving).

In an event driven architecture:

  • the events are facts.
  • Events are the notifications (trigger) for transition.

In the visual flow, the nodes are the transformation, not the state of the entity as in a process model.

The events are clustered to create a chain of transformations based on their relationship. They are not grouped based on the state of the entity as in a process oriented model.

The model start with an event and describes all transformations on the entities. ie the model system's behavior is described in terms of a sequence of transformation (not in a sequence of state).

Activities are not modeled directly as in a life-cycle model. Rather they are implied by scheduling an event for the point in simulation time they will be finished.

The main entry in a event-driven system is called an event loop.

3 - Steps

3.1 - Identify the relevant entities

To build an event-based model, we must identify all relevant objects and their properties.

3.2 - Describe the behavior of the system

  • Description of Behavior (Events)
  • identify the events that “drive” our model (ie the events that change the state of the system)
  • Write down the types of events that occur in the system
  • For each event type, specify the transformations and interactions of the relevant entities, e.g. by drawing a UML activity diagram. Example

4 - Documentation / Reference

Data Science
Data Analysis
Data Science
Linear Algebra Mathematics

Powered by ComboStrap