Study of Multi-Behavior Agent Systems for Supply Chain Planning In Bearing Industry

Abstract

For any industry or company, to be a competent in market needs increase in performance in all possible ways. Quick response to market and sufficient production as per requirement is the only way to increase the returns, which require critical planning and production systems. For this co-ordination between business units or workstations is essential. Many of the executive managers in industries has to instruct only the task flows to its subordinates, that is a single straightforward production planning process is followed as executed from top level of organization; their capacity as per their education level is not utilized more than 10%. The use of multi-agent system allows physical distribution of the decisional system and procures a hierarchical organization structure with decentralized control that guarantees the autonomy of each entity and flexibility of network. Our study focuses on managers/partners that adapt together their local planning process to face different requirements of supply chain environment using different planning strategies, when decisions are supported by distributed planning systems. The agent based system has the advantage of making collaborative management of disturbances in supply chain as the agents has the advantage of making autonomous decisions in a distributed network. Because each partner can choose different behavior and all behavior has an impact on the overall performance, it is difficult to know which is preferable for each partner to increase their performance. Thus, in this paper study of Multi-behavior planning agent model is done using different planning strategies when decisions are supported by distributed planning system. Keywords: Multi-agent, agent-based planning systems, collaborative supply chain planning. I. INTRODUCTIONThe bearing industry is highly distributed, with many production units interacting in all activity levels. Traditionally, companies used to view themselves as separate entities and did not devote efforts to collaborate with other echelons of the extended enterprise. The main problems of this industry lies in the large amount of stochastic disturbances in many aspects of the supply chain, mainly due to the highly heterogeneous aspect of the resource, uncertain process output, production of co-products and by-products, price variation in the spot market, and demand variation in commodity markets. Collaboration among entities in the supply chain can have a great impact on the system performance. The various OR techniques were applied to improve planning, coordinating, transporting between the manufacturing units and workstations within units. But all are centralized controlled and this may result in much of the system being shut down by a single point of failure, as well as plan fragility and increased response overheads. Agent technology provides a natural way to address such problems, and to design and implement efficient distributed intelligent manufacturing systems. Agent technology has been recognized as a promising paradigm for next g...