This research is contributing to the development of a Scope, Cost and Effort Estimation Framework for Service Oriented Architecture (SOA) projects.
Current approaches to costing SOA projects are very limited and have only been applied to specific types of SOA such as Service Development or SOA Application Development. The use of conventional estimation approaches such as function points for the development of large systems have been found to not be useful as there are issues such as determining the ‘boundary position’ (what is within the scope of the software to be developed and what is outside the system) and determining the unit of size in order to apply the approach.
As a result there are no methods, tools, cost models, or frameworks that cover scope, cost and effort estimation for the collection of SOA projects outlined in the figure below. Some research is being done that exclusively examines the Service Mining project type at the Software Engineering Institute.

Many SOA projects face risks of blow-out of cost, and effort and many such projects do not deliver on time. Research in the Service Scoping and Estimation area aims to examine the state of the art in industry and government sectors on SOA projects with a focus on determining what exactly is done in terms of scoping such projects, what is the best practice in estimating the effort and costs involved in such projects and examining not only the technical work involved but also the social/cultural/ organisational work that has to be carried out on these projects. The research challenges include:
It is envisaged that large parts of the scope, cost and effort estimation framework will be implemented in a tool (or collection of tools) that can be used by organisations. The framework can be used on an ongoing basis by an organisation once a project commences to verify and keep track of the cost and effort estimates. In this way any potential for cost blow-out can be highlighted early. There are three main uses envisaged for the framework:
There is a need for an empirical validation of the Framework to determine the success of the research. This validation will depend on successful use and deployment of the Framework within organisations. Organisations should be able to do a more comprehensive scoping and estimation of their projects that includes factors that have been empirically validated to be critical. As a result of adopting the Framework these organisations should be in a position to rely less on SOA estimation experts to perform estimates. It is planned to trial the Framework (in part or whole) on various SOA projects to scope and estimate cost and effort for these projects. There is also a need to monitor the cost and effort spent on these projects over time and to determine if the estimates were valid. In order to do this there is a need for partner organisations that will allow us to field trail the Framework and be involved in and observe their SOA project from inception to completion which could span over months or years.
An overview of the initial framework is provided in the following paper: