Formal requirements models from domain knowledge
David Furber, S. Green, David L. Bolton, Sara Jones, David Till · 1992
While large sums of money are being spent on developing software, studies indicate that this money is largely wasted: in one case less than 2% of the money spent on nine software projects resulted in software that met its requirements. Research has shown that a lot of the money is spent at the beginning of the software development cycle. This observation seems to be reinforced by indications that software errors are relatively cheap to detect and fix during the requirements and planning stage, but much more expensive later in the development cycle. Considerations such as these have resulted in an increase on requirements engineering. The authors attempt to use and reuse knowledge about a domain to improve the processes of eliciting requirements, analysing requirements, and creating specifications, for applications in the domain. >