Testing - State Transition Testing

State Transition Testing

State Transition Testing can be used on any system that has stable states, inputs that cause it to change state (Events), and outputs that are generated as the result of a state change (Actions). An example would be a TV set where the set remains off until switched on (Event), and will display the selected channel (Action) until the remote control is used to trigger a change to a different channel.

The technique states that there will be a finite number of unique and identifiable stable states, along with the Events (triggers) that cause a state change and the Actions (outputs) that are generated as a result of a state change. The technique allows the States, Events and Actions to be represented diagrammatically on a State Transition Diagram and also allows the production of a State Table that will show all combinations of States and Events to allow testing of both valid and invalid transitions (this is to check that no transition occurs in response to an invalid trigger for that state).

Technique steps:

• Identify the stable States, Events and Actions.
• Produce a State Transistion Diagram
• Produce a State Table
• Define tests for all valid and invalid transitions as detailed on the State Table
• Define tests to achieve the required Chow coverage from the State Transition Diagram based on the risks.