Using Multi-Agent Based Simulation in an IDSS for Evaluating Performance in Flow-shop Manufacturing System

摘要

This paper presents an approach which takes into account robustness and flexibility of a real time scheduling, by integrating agents into an IDSS. The proposed system mainly includes six components: Resource Agent (RA), Planning Agent (PA), Performance Evaluation Agent (PEA), Database Management Agent (DMA), Rules and Criteria Selection Agent (RCSA) and User interface Agent (UIA). The multi-agent simulation is used to allow agents to cooperate with an intelligent behaviour, and to coordinate their goals and action plans to solve a problem. We use UVA methodology to calculate the production costs. This method provides the enterprise with new information on its performances, the profitability of its customers, markets, product, which will generate decisions in all business functions for a permanent progress. One objective of this work is to demonstrate that the UVA method is not an accounting method of repartition and that it is built on the choice of one measuring unit and one specific analysis approach. Experimental results demonstrate that the proposed IDSS could implement effective production control decision-making for solving the flow-shop manufacturing system. The study reports the basic design principles of the system as well as details of the application.