A concept of quality assurance for metrics in CASE-tools

Abstract

The quality of software has become more important to software companies in the past years. Software measurement is one of many approaches that is used to check the quality of software [18]. This may involve measuring some attributes of a software product or a software process and comparing these measurements to each other or to some desirable level. A software metric is "any type of measurement that relates to a software system, process or related documentation" [7]. Software metrics can help to improve the quality of the produced software. However, metrics and metrics tools are still not used in most software companies -- for example in the 3 companies where we cooperated in last 5 years. One reason is that there is lack of knowledge about metrics and hence software metrics are still unknown or difficult to use for some developers, and software measurement is still time-consuming for managers. Another reason is that, good metrics tools are still expensive for small and middle companies. In our opinion the effectiveness of metrics can be improved by simple organizational expedients. In this paper we present new/adapted requirements on metrics in CASE tools to define flexible product quality models. For this quality model we tried to use some standard techniques, e.g. metrics suite, metrics visualization or metrics filtering to show how metrics in CASE tools can be defined and how it can benefit different people who are involved in a software deployment.