Research Article

Agent-Oriented Software Engineering Methodologies: Analysis and Future Directions

Table 1

Comparative analysis of Gaia and Tropos methodologies.

MethodologyStrengthsLimitationsApplication domain

Gaia(i) Directs designers to a well-defined design for the MAS(i) Agents cannot share common goals(i) Application of agent methodology in healthcare information systems [46]
(ii) Model and copying with the features of complex systems(ii) It does not handle systems while acting as an actor(ii) Multiagent system vulnerability detector for a secured E-learning environment [66]
(iii) Is easy to apply(iii) The coherence between the protocols is rather weak(iii) Multiagent-based transformer condition monitoring [67]
(iv) Components of the system may enter or leave at runtime in open or dynamic systems
(v) It addresses goals, but verification of goals is still out of scope
(vi) It does not UC scenarios [23]

Tropos(i) Suggests methods and tools for automating models transformation(i) It is not designed to support a certain type of software.(i) A process for developing adaptable and open service systems: Application in supply chain management [30]
(ii) Focuses on the preliminary stage of requirements(ii) Developed software agents designed for accomplishing special goals are not Supported by Tropos methodology(ii) Using Tropos to model agent-based architectures for adaptive systems: a Case study in ambient intelligence [39].
(iii) Uses actors and goals as basic concepts [68].(iii) Does not UC scenarios [23].(iii) Developing a decision support system for integrated production in agriculture [52].