Analytical techniques for focussing software research

For all the advances in the management of software research and development over the last few years, it is never easy to ensure that complex programmes of research are efficiently directed and assessed. There are a number of inherent difficulties that contribute to this, for instance-the complexity of modern technology, the abstract nature of research and the need to work in an area which is, by definition, largely unexplored. This paper discusses two techniques aimed at easing the evaluation of complex projects. Both techniques are designed to ensure that critical decisions and their dependencies are explicitly addressed. The effectiveness of the techniques is illustrated by their application to research in software engineering.