SFT5050

Software metrics and reliability

C Mingins

6 points
* 4 hours per week
* First semester
* Caulfield

Objectives At the completion of this subject, students should have gained an understanding of the formal aspects of software measurement that enables them to evaluate existing software metrics critically; evaluate the suitability of existing metrics for object-oriented requirements; evaluate and construct software reliability models; and develop measurement tools that will collect data representing the application of stated design theories to an existing system.

Synopsis Seminar discussion will include the following topics: review of principles of measurement; formal approaches to software measurement; measuring software `quality'; metric validation and verification; data collection; measures in object-oriented systems, complexity measures, reuse metrics, design measurement; stating dependability requirements; measuring the reliability of software. Students will be expected to work in teams to develop a set of metric tools to measure aspects of an object-oriented project development.

Assessment Participation in classroom discussion: 25%
* Practical assignment: 75%

Prescribed texts

Fenton N E Software metrics. A rigorous approach Chapman and Hall, 1992

Recommended texts

Denvir T and others (eds) Formal aspects of measurement Springer-Verlag, 1992
Grady R B and Caswell D L Software metrics: Establishing a company-wide program Prentice-Hall, 1987
Kitchenham B A and Littlewood B (eds) Measurement for software control and assurance Elsevier, 1989
Kyburg H E Theory and measurement CUP, 1984
Musa A I and Okumoto K Software reliability: Measurement, prediction, application McGraw-Hill, 1990