Multi Agent Based Framework for Structural and Model Based Test Case Generation

Abstract

Software testing is done to find errors and fix them to improve quality. Software testing is a laborious and time–consuming work which spends almost 50% of the software system development resources. The number of test cases required to develop error-free software is high and when done manually the process of test case generation is time consuming and error prone. As systems are increasing in complexity, more systems perform mission-critical functions and the dependability requirements such as safety, reliability and security are vital to the users of the system. The competitive market place forces companies to define new approaches to reduce time-to-market as well as development cost of the systems. To increase the effectiveness and efficiency of the testing process and to reduce the overall development cost for software system, new approaches are required for test automation. The automation of test case generation is the most important aspect of test automation. No powerful test data generation tools combining structural and functional testing are commercially available today. The objective of the proposed system is to automate the process of test case generation for both structural and model based testing by deployment of an agent based framework. This framework automatically generates an optimized Test Suite for system under test that can assist the testers and developers and enable them to correct the errors in earlier stages of software development. The deployment of agent based framework reduces considerable amount of execution time when compared with the system without agents.