Model based approach to detect emergent behavior in multi-agent systems

Abstract

Multi-agent systems (MAS) are efficient solutions for commercial applications such as robotics, business commerce applications, information retrieval and search engines. In MAS, agents are usually designed with distribution of functionality and control. Lack of central control implies that the quality of service of MAS may be degraded because of possible unwanted behavior at runtime, commonly known as emergent behavior. Detecting and removing emergent behavior during the design phase of MAS will lead to huge savings in deployment costs of such systems. Effective and efficient design validation of MAS requires the development of systematic and automated methodologies to review MAS design documents. Although the increasing demand for MAS in the software industry has led to the development of several Agent Oriented Software Engineering (AOSE) methodologies, the AOSE methodologies usually do not fully cover monitoring and testing. In this paper, a technique to help MAS developers verify, test and monitor MAS design is introduced. This method uses MAS analysis and design artifacts created by the MaSE AOSE methodology. In this technique, design artifacts of MaSE are converted to scenario-based specification, which is very similar to UML's sequence diagrams. Then the specifications are used to analyze the system for validating the design of MAS and ensuring the lack of emergent behavior.