The effect of test case design in software testing bots

Abstract

Traditional approaches of testing in software development include running the test cases on a software component, referred to as unit testing, which usually only tests a specific part of a component, as opposed to testing the whole flow of the system (end-to-end testing). Test bots are software automation tools that help improve the system testing via automation, which is beneficial for development teams as the test bots help decrease the amount of time spent on testing. As development projects become larger, it is important to focus on improving the test bot’s effectiveness. The test bots run a set of test cases that check whether the system under test meets the requirements set forth by the customer. This thesis uses a case study approach to investigate how test case designs can affect the test bots, and by using the findings gathered from the study, we aim to create a guide for test design schema for such bots. Furthermore, this study aims to find how the software testing practices in an IT company can differ from what the literature presents. We identify the main challenges when using test bots in the automotive industry and a guideline is composed of seven steps to aid stakeholders in designing tests where test bots are part of the testing cycles.