| Techniques |
Applicability |
|
TSR
|
Industry Motivation
Industry Evaluation
Industry Author
|
| Experiment subject(s) |
Industrial Partner |
Programming Language |
| SCAN tool from ABB
Industrial proprietary, unclear scale |
ABB (India/USA) |
Unclear |
| Effectiveness Metrics |
Efficiency Metrics |
Other Metrics |
|
Selection/reduction count/percentage, Testing time
|
|
|
| Information Approach |
Algorithm Approach |
Open Challenges |
|
Requirements-based
|
|
generation with similar approach; consider time of each test case.
|
Abstract
Test suite minimization ensures that an optimum set of test cases are selected to provide maximum coverage of requirements. In this paper, we discuss and evaluate techniques for test suite minimization of evolving software systems. As a case study, we have used an industrial tool, Static Code Analysis (SCAN) tool for Electronic Device Description Language (EDDL) as the System Under Test (SUT). We have used standard approaches including Greedy, Greedy Essential (GE) and Greedy Redundant Essential (GRE) for minimization of the test suite for a given set of requirements of the SUT. Further, we have proposed and implemented k-coverage variants of these approaches. The minimized test suite which is obtained as a result reduces testing effort and time during regression testing. The paper also addresses the need for choosing an appropriate level of granularity of requirements to efficiently cover all requirements. The paper demonstrates how fine grained requirements help in finding an optimal test suite to completely address the requirements and also help in detecting bugs in each version of the software. Finally, the results from different analyses have been presented and compared and it has been observed that GE heuristics performs the best (run time) under certain conditions. Copyright © 2019 by SCITEPRESS - Science and Technology Publications, Lda. All rights reserved.