Journal of Computer Science

An Ontology Driven, Procedural Reasoning System-Like Agent Model, For Multi-Agent Based Mobile Workforce Brokering Systems

Arash Mousavi, Md. Jan Nordin and Zulaiha Ali Othman

DOI : 10.3844/jcssp.2010.557.565

Journal of Computer Science

Volume 6, Issue 5

Pages 557-565


Problem statement: It has been proven that Believe, Desire, Intention (BDI) agent architecture, performs suitably in dynamic and unpredictable environments. Although BDI architecture has been formulated rigorously, implementing BDI architecture is not as straightforward as it has been promised. Nevertheless, the preeminent implementation of BDI architecture is Procedural Reasoning System (PRS) and Jadex is a Java-based platform that successfully applies PRS BDI architecture to implement BDI agents. However, Jadex utilizes XML format to represent plans, beliefs and goals which are the core of the PRS BDI agents. Although XML is a suitable format for exchanging data, it does not add any semantic to the data that it represents. Approach: In contrast, in our proposed Ontology driven PRS like model (O-PRS), we used Ontology with OWL format to represent Believes, Plans and Events. Our Ontology had been designed particularly for implementing BDI agents which are used in Mobile Workforce Brokering Systems (MWBS), a multi-agent system that automated the process of allocating tasks to Mobile Workforces. Results: This research was an endeavor to equip the PRS model with Ontology and semantic knowledge representation and reasoning capability. In this study firstly, our proposed model was formulated, then the model was been examined using a simulation tool (MWBS-SIM) which simulated MWBS in its Initialization phase. We used JADE to implement the Agents, Protégé OWL to create the Ontology and Jena as a semantic web platform by means of which MWBS is able to connect to the Ontology and apply the O-PRS model. Conclusion: The result of our simulation showed that O-PRS is a feasible approach. Furthermore, Ontology provided better means for knowledge sharing and navigation therefore, multi-agent systems constructed using O-PRS, were more generic and more adaptable in comparison with those that use XML.


© 2010 Arash Mousavi, Md. Jan Nordin and Zulaiha Ali Othman. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.