Abstract
Despite the large supply of methods and tools for cost estimation, estimating the costs of a software development project remains a non-trivial activity. Research [Heemstra 1989, Mermaid 1989] has shown that the accuracy of such tools is low. It has also been shown that only a limited group of organizations uses systematic methods for drawing up a cost estimate for project-based software development. For many years, various lines of industry (for example the building industry and the catering industry) have used experience figures when drawing up cost estimates for projects. Cost estimates for software projects often lack such a basis, partly because there are no adequate tools for recording and analysing historical project data. This paper presents an analysis of the problems in the field of software cost models and describes the MERMAID approach to cost estimation. The MERMAID approach makes intensive use of local historical project data and is applicable in all sectors where project data can be collected.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call