Combined Architecture for Early Test Case Generation and Test suit Reduction

Model based combinatorial testing is the most effective and efficient method of systematic interaction testing. In this multiple variables can interact with test model in the form of combination and each set is considered as a test pairs. It is used mainly for reducing the test pairs size, complexity and time of test generation. Accomplishment of this reduction task is the most vibrant and tedious activity because this consist of test data extraction, boundary analysis, path measurement analysis, condition coverage etc. The problem with Model based combinatorial testing is optimization of cost and effort which can only be achieved by early test case generation methods. For this we are using UML diagrams from design phase. For reducing the test size and complexity we have proposed a new MBTGA framework and an MTGIPO. In this paper we focus our research on empirical study and result obtained by developed framework. We are also proposing a new UML design data extraction (DDE) algorithm for getting the useful information from a given UML diagram as input.