Designing for Planned Emergence in Multi-agent Systems

Abstract

We present an approach for designing organization-oriented multi-agent systems (MASs) to allow improvisation at run time when agents are not available to exactly match the original organizational design structure. Working with system components from an existing MAS organizational meta-model, OJAzzIC, the approach sets out five stages for the design process. We illustrate the design approach with an incident response scenario implemented in the Blocks World for Teams (BW4T) environment, and show how agents at runtime can improvise -for example they can adopt tasks even if those tasks do not precisely match a predefined role.

Designing Multi-Agent System Organisations for Flexible Runtime Behaviour

“…In this section, the design methodology, an adaptation of the O-MaSE methodology [30], is briefly outlined for adopting the OJAzzIC model to create a MAS. The design methodology is described in more detail elsewhere [74,78].…”

“…The OJAzzIC meta-model as described conceptually elsewhere [21,22,74] includes a specification of agents in an organisation, an organisational structure, and a set of behavioural policies and guidelines. Policies specify how multiple organisations can be created to form a distributed, coordinated system of interconnected agents and organisations and define explicit agent obligations.…”