Job shop production can be characterized by some unplanned event like re-entrant jobs at a certain stage of the production process. In this case a dynamic allocation management of the resources involved (machines, transporters, etc) can be key factor of success in optimizing some production parameter. The present work studies a multiple objective job shop in which the constraint that a workpiece visits a machine only once is relaxed. The production shop floor has been structured with a Multi Agent System (MAS) able to front dynamically these type of events. A benchmarking analysis is provided to compare the solutions found with other two main dynamic MAS-based systems present in literature.