An investigation into evolving support for component reuse

Abstract

It is common in engineering disciplines for new product development to be based on a concept of reuse, i.e. based on a foundation of knowledge and pre-existing components familiar to the discipline's community. In Software Engineering, this concept is known as software reuse. Software reuse is considered essential if higher quality software and reduced development effort are to be achieved. A crucial part of any engineering development is access to tools that aid development. In software engineering this means having software support tools with which to construct software including tools to support effective software reuse. The evolutionary nature of software means that the foundation of knowledge and components on which new products can be developed must reflect the changes occurring in both the software engineering discipline and the domain in which the software is to function. Therefore, effective support tools, including those used in software reuse, must evolve to reflect changes in both software engineering and the varying domains that use software. This thesis contains a survey of the current understanding of software reuse. Software reuse is defined as the use of knowledge and work components of software that already exist in the development of new software. The survey reflects the belief that domain analysis and software tool support are essential in successful software reuse. The focus of the research is an investigation into the effects of a changing domain on the evolution of support for component-based reuse and domain analysis, and on the application of software reuse support methods and tools to another engineering discipline, namely roll design. To broaden understanding of a changing domain on the evolution of support for software reuse and domain analysis, a prototype for a reuse support environment has been developed for roll designers in the steel industry.