Abstract. We introduce a distributed architecture to endow multi-agent systems with a social layer in which norms are explicitly represented and managed via rules. We propose a class of rules (called institutional rules) that operate on a database of facts (called institutional states) representing the state of affairs of a multi-agent system. We define the syntax and semantics of the institutional rules and describe a means to implement them as a logic program. We show how the institutional rules and states come together in a distributed architecture in which a team of administrative agents employ a tuple space (i.e., a kind of blackboard system) to guide the execution of a multi-agent system. 1

The use of Multi-Agent Systems (MAS) in health-care domains is increasing. Such Agent-mediated Medical Systems are designed to manage complex tasks and have the potential to adapt gracefully to unexpected events. However, in that kind of systems the issues of privacy and security (in the access to patient records), safety and soundness (of the individual agent behaviours and the multiagent system as a whole) and trust (among heterogeneous agents and among users and agents) are particularly sensitive. An additional issue is that health care systems are highly regulated by regional, national and European regulations and policies. Therefore several normative contexts (European, national, regional) should be taken into account while designing agent-mediated health care systems. In this paper the less explored normative problem of the application of agents in Health Care will be presented. The impact on both the individual agents and the agent platforms will be discussed. On the individual agent side, Normative Agents (agents whose behaviour is guided by collections of norms) will be presented. On the agent platforms side, the concept of Electronic Institutions (as normative environments to increase trust and ensure proper agent behaviour) will be presented, along with some methodological considerations, the impact on implementation and some of the technologies needed to be developed in the future. 1

Abstract. A major challenge in normative multiagent systems (NMAS) is how norms can be effectively applied to their entities and easily managed. These tasks are arduous because norms are usually written for general purposes, hindering a more precise regulation. In this paper, we go through an operational approach to implement regulative norms in multiagent systems. In our approach, abstract norms are embodied with domain values according to agents’ current contexts. Thus, we try to bridge the gap between the theoretical work on norms and practical normative systems providing a solution to implement concrete regulative norms in NMAS. The rationale of the approach is described via a motivating example from the domain of international supply-chains.