As involved in large turn-key projects, the company experiments the more and more huge need for:
1. providing managers with historical information to use in prediction and control of software projects;
2. guaranteeing independent quality assurance through an objective assessment of both the properties of developed software and the software development process;
3. providing indicators of the effectiveness/efficiency of the company's quality system.
In the evolution of the company's quality systems, already certified compliant to ISO 9001, the set up of a software product and process metrics database is crucial. The objective of the experiment is not only to select the most appropriate metrics, but also to identify effective and efficient procedures for collecting and storing meaningful and complete data.
The achievement of the objective above requires the improvement of the data collection and analysis procedures involving the production and quality assurance functions. The Goal-Question-Metric paradigm is used for the systematic definition and establishment of the measurement programme. The experiment will use as baseline the CNPAIALP project - a large turn-key project devoted to the development of the information system for an Italian insurance body. The project team involves 20 people.
EXPECTED IMPACT AND EXPERIENCE
Engineering expects to take advantage from an invaluable information source for identifying improvement needs and validating proposed innovations. A very important organisational outcome will be a more co-operative and less conflicting relationship between quality assurance and production people.