Test Case Prioritization Using Fuzzy Logic Based on Requirement Prioritizing

Boolean expressions are popularly used for modelling decisions or conditions in specifications or source programs and they are very much prone to introduction of faults. Even for a Boolean expression with few numbers of literals the possible number of test cases can be quite large. Boolean expressions with n variables require 2 n test cases to distinguish from faulty expression. In practice, n can be quite large and there are examples of specification having Boolean expressions with 30 or more variables. To test a system based on Boolean specification in limited time, it is not possible to execute all test cases so prioritization is required which leads to early fault detection in testing life cycle. There are various testing strategies for generation of test cases for Boolean specifications like MUMCUT, which generate fewer test cases then 2 n with high probabilities of finding errors but their prioritization are not considering the criteria from user’s perspective. We have proposed a new approach which prioritizes test cases based on requirement prioritization. Our aim is to find the severe faults from user’s perspective early in the testing process and hence to improve the quality of the software .This paper considers method for assigning weight value on the basis of factors which generates the criteria for test case prioritization for Boolean Specifications. These factors are: Business Value Measure (BVM), Project Change Volatility (PCV), and Development Complexity (DC). Priority is assigned to test cases based upon these factors using fuzzy logic model.