What is TPT (time partition testing) test methodology?

TPT test methodology or Time Partition Testing is quite a systematic one and is used for carrying out automated testing of software systems and applications. This tool is also used to verify the data flow programs and embedded control systems as well. 

About Time Partition Testing

- Time partition testing methodology has been designed exclusively for the validation and verification of the embedded systems whose input and output data can be represented as signals. 

- Also, this tool is used for testing the continuous behavior of the software systems and applications. 

- Many of the controls systems belong to this category of software systems and applications.

- One of the characteristic feature of such systems is that they interact very well when interlinked with the real world environment. 

- There is a need for the controllers to closely observe the environment and react accordingly towards it. 

- There exists an interactive cycle within this environment in which these systems are supposed to work. 

- This cycle is subject to the temporal constraints. 

- Testing such systems means simulation and checking of their timing behavior. 

- The time partition testing makes use of model based testing rather than using scripts like the traditional functional testing methods. 

- Time portioning testing methodology makes a combination of the graphic and systematic modeling techniques for running the test cases fully in a variety of environments and need to be evaluated automatically. 

- The below mentioned 4 test activities are covered by time portioning testing methodology:

1. Test case modeling
2. Automated test documentation
3. Test management
4. Automated test execution in a number of environments
5. Automated test result analysis including automated test assessment

- In this methodology, the special state machines and time partitioning methodology is used for modeling the tests. 

- Hybrid automation can be done for all the tests belonging to an SUT (system under test).

- Tests come with a logical phase sequence. 

- The logical passes are somewhat similar for all the tests and are represented by the states of the finite state machine. 

- The transitions that occur between the phases of a test are modeled by trigger conditions. 

- Each transition is characterized by different variants. 

- These individual variants are made in to unique combinations which in turn are used for modeling individual test cases. 

- Natural language texts, since they support demonstrative as well as simple reading for programmers and non – programmers alike, are known to become a part of the graphics. 

- There are certain substantial techniques which help in modeling the complex test cases via intuitive and graphic modeling:

1. Parallel state machines
2. Hierarchical branching state machines
3. Conditional branching
4. Re-activity
5. Signal description
6. Measured signals
7. Test step list

- Graphics hide the complexity of a test. 

- Test step lists (also known as direct definitions) constitute the lowest level of the signal description. 

- The last technique i.e., the test step list implies a certain number of predefined actions which are executed sequentially and therefore are ordered chronologically. 

- This technique also holds good for the test cases which can be described normally in a sequential manner which can be chosen from the following:

1. Channel set i.e., setting a signal
2. Ramp channel or ramp signal
3. Set parameters
4. Compare
5. Wait
6. Other conditional values like if, else if, else values and so on.

- In addition to this, these sequences can be modeled in combination with the other models. 

- Direct definition method is used for specifying the test-lets which involves an ordered list of equations and signal definition. 

- Signals can be imported via a function wizard or manual signal editor. 

- However, TPT remains exclusive for testing the reactive behavior of the embedded systems.