AN ECOGENIC MULTI-AGENT FRAMEWORK FOR STEP-COMPLIANT MANUFACTURING: AN ARTIFICIAL IMMUNE SYSTEM BASED APPROACH

摘要

Over the years, G codes have been extensively used by the CNC machine tools and are now considered as a bottleneck in making these machines reconfigurable and interoperable. This paper explores the application of distributed artificial intelligence methods in STEP-compliant manufacturing to realize the interoperability in such systems. An ECOGENIC multi-agent system has been conceptualized that will offer an integrated infrastructure for building distributed computing systems and provide a better option for modeling the interoperable manufacturing environment with simulating behavior. The present system facilitates a highly configurable interface for modeling the manufacturing environment, where different shop floor activities are performed by the agents with desired capability in their skill stream. The present framework utilizes the STEP-NC standards and provides functionalities, like bi-directional flow of information, in CAD/CAM, and replaces traditional tool movement description languages with complex object oriented data structures. The primary focus of this paper is the proposed ECOGENIC multi-agent architecture for CAx process chain, while the secondary aim is to demonstrate the working mechanism of the ECOGENIC process planner in the operation sequencing problem and solving it by an artificial immune system (AIS) based approach. A comparative evaluation has been made with the genetic algorithm which validates the AIS as a viable and robust solution methodology for such problems.