Development of complex systems often requires building a large number of models with many interconnections and dependencies among them. The success of a project can be compromised by cognitive overload or limits of developers, who might miss relationships between elements of the models. Developing Multi-Agent Systems (MAS) is a typical example of where this may occur. Despite of its potential, this technology has not yet been widely adopted by industry due to its complexity and frequent errors in modelling activities. These errors typically propagate to later phases of the MAS development lifecycle, becoming costlier to fix and then lowering the quality of the final product. Early validation of MAS models can prevent rework efforts or building a system that is non-compliant with the client's specification. In this paper we propose a process to support developers in modelling tasks using ontologies to validate and improve the quality of requirement analysis models as they are being developed and at the same time bridging the traditional gap between developers and clients. The proposed ontology-mediated validation is easily applicable to other kinds of architectures, however we illustrate this for MAS development as its complexity justifies any additional cost associated with applying it.
Link to publisher version (URL)