An ontology-based multi-agent virtual enterprise system (OMAVE): part 2: partner selection

Abstract

A virtual enterprise (VE) is a collaboration model between multiple business partners in a value chain. The VE model is particularly feasible and appropriate for small- and medium-sized enterprises (SMEs) and industrial parks containing multiple SMEs that have different vertical competencies. The VE consortium’s success highly depends on its members. Therefore, it is crucial to select the most appropriate enterprises when forming a VE consortium. In this study, a new multi-agent hybrid partner selection algorithm is developed for application in the development of an ontology-based multi-agent virtual enterprise (OMAVE) system. In this platform, the agent’s interactions are supported by agent ontology, which provides concepts, properties and all message formats for the agents. Different types of agents collaborate and compete with each other so that unqualified or inefficient enterprises are eliminated from the enterprise pool. Only the remaining enterprises would be allowed to enter the negotiation process and propose in the bidding. The agent-based auctioning platform is coupled with a fuzzy–AHP–TOPSIS algorithm to evaluate partners based on their proposals and background. Accordingly, the winning enterprise for each task is identified and the whole project can be accomplished by assigning tasks to the responsible partners. To test and verify the functionality of the developed OMAVE system, a sample module using OMAVE applications and tools was manufactured. The last section of this paper presents the results of this case study, which validate the applicability of the proposed technique.

Keywords:

• virtual enterprise
• partner selection
• multi-agent systems
• bidding

Acknowledgements

This study has been funded by the SAN-TEZ program, project no. 00979.stz.2011-12, of the Turkish Ministry of Science, Technology and Industry. The authors are sincerely thankful for the continuous support of the OSTIM OIZ management. The authors also wish to thank Fatih Arikan for his valuable contributions and efforts in modelling and creating the agent-based platform on JADE.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. OSTIM Technology A.S. was originally established to enhance the collaboration between universities and industries, with contributions from academics at universities and from companies at the OSTIM Organized Industrial Zone.

Additional information

Funding

This study has been funded by the SAN-TEZ program, project no. 00979.stz.2011-12, of the Turkish Ministry of Science, Technology and Industry.