METRIX: a tool for software-risk analysis and management

In order to improve the production process and to guarantee the quality of the manufactured products, more and more sophisticated data is needed. Also mathematical models are needed to exploit them: these models have to be useful for prediction and easy to interpret so that remedial actions may be taken, as early as possible, in order to control and optimize the production process. These models enhance their efficiency when integrated into decision support tools. The previous remarks, applied to the software development process, have led the authors to study an efficient modeling technique to model software quality and to automate it in the METRIX tool. The paper presents together the modeling technique (i.e. optimized set reduction), the tool that supports it (i.e. METRIX), and the global strategy in which the tool is to be used. The efficiency of the modeling technique is demonstrated by results obtained for the evaluation of "high risk" components in several Ada systems. However the technique is not limited to Ada or any specific development environment. >