Invited presentation II: Experiences in applying model based system testing generation

0Citations
Citations of this article
7Readers
Mendeley users who have this article in their library.
Get full text

Abstract

The goal of this presentation is to illustrate the benefits of using an automated, model-based approach for improving system test design and generation. Our approach, TDE/UML, automatically generates system tests from behavioral models of an application using the Unified Modeling Language (UML.). TDE/UML builds on and combines existing techniques for data coverage and graph coverage. We focus here on the results of applying TDE/UML in diverse Siemens projects: its cost benefits and its fault detection capabilities. System testing, which ensures the functional compliance of an application with its requirements, is a well-defined process within Siemens. However, in many cases, it remains a manual process. Test designers typically derive their system input and expected output information from a variety of sources including textual use case specification and business process rules. They then create a set of test procedures comprising of individual test steps, which are executed manually by test executors against the system under test. Whenever an automated test environment is available, these test executors are also responsible for translating these textual test procedures into executable test scripts. To automate and formalize this process as much as possible, our research uses model based system test generation. Our approach makes use of UML Use Case diagrams to describe the relationship among the diverse use cases implemented by the system and the actors who interact with the system through those use cases. UML activity diagrams are used to model the logic captured by a single use case and between use cases. The set of activity diagrams represents the overall behavior specified for the system and it is the basis for testing the different functionalities and business rules described in the use cases specification. Tests are automatically generated from those models following three phases. During the first phase, a test designer manually annotates the UML models based on SCR Test Specification Language (TSL), which implements the category-partition methodology. In the second phase, the test generation tool (TDE/UML) automatically creates a set of textual test procedures (test cases) or executable test scripts. In the third phase, a test executor runs these against the system under test using a commercial UI testing tool. © Springer-Verlag Berlin Heidelberg 2005.

Cite

CITATION STYLE

APA

Vieira, M. (2005). Invited presentation II: Experiences in applying model based system testing generation. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 3713 LNCS, p. 430). https://doi.org/10.1007/11557432_32

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free