Abstract
In recent years, it has become difficult to continue supplying system quality while meeting the increasing demand for more rapid development at lower cost. Even if in-depth quality control measures are carried out in the early phases of the system development life cycle, there is no other way to verify the reliability and quality of the system than putting it to the test. However, in order to save costs and resources the significance of 'in house' testing tends to be disregarded, and testing is assigned by SI vendors to their Business Partners (hereinafter, "BP"). Consequently this raises another issue that SI vendors may not supply the quality due to the lack of their control on defining test specification and verification points. This paper discusses the following three measures, which were put into practice to handle this issue in a mission-critical system development project for a financial customer. 1) Fostering a lead testing team skilled in customer's business procedures, 2) Risk avoidance by strategic testing plan, 3) Conducting the test without entire dependence on BP. Namely, the lead testing team concentrated on setting up and operating the testing environment and controlling the process of the test, while BP concentrated on quality assurance activities. This project achieved a result with a high level of customer's satisfaction and early stabilization of the launched system. Furthermore, the knowhow practiced in this project was organized as an industry-specific test framework, so that it could be practicably reused for other projects.
