Paper

Proceedings of the Second International Intelligent Logistics Systems Conference 2006

A MULTI-AGENT APPROACH FOR ANALYSING MATERIAL FLOW IN A MANUFACTURING SUPPLY CHAIN Ruhul Sarker*, Graham Freeman*, Sami Kara**, Berman Kayis**, Tapabrata Ray*** and Hussein Abbass* University of New South Wales, Australia * ADFA School of ITEE and ***ADFA School of ACME, Northcott Drive, Canberra 2600**School of Mechanical and Manufacturing Engineering (Sydney) r.sarker@adfa.edu.au ABSTRACT In the last decade, supply chain operations have received tremendous attention in manufacturing and business sectors due to an increasingly challenging marketplace. This paper presents a multi-agent approach for analysing material flow in a manufacturing supply chain under information sharing. The model is capableof handling complex networks with many tiers, each tier with many business units and complex interactions among them. We have discussed the multi-agent architecture and run simulations for analysing the operational aspects under both deterministic and stochastic demand. This will allow companies to quantify different interacting parameters in the supply chain and support improvement inoperations. Key Words: Supply Chain, Multi-Agent Approach, Material Flow. 1. INTRODUCTION A supply chain is “an integrated process wherein a number of various business entities (i.e., suppliers, manufacturers, distributors, and retailers) work together in an effort to: (i) acquire raw materials /ingredients /components, (ii) convert these raw materials /ingredients /components into specified final products,and (iii) deliver these final products to retailers” (Beamon,1998). This chain is traditionally characterized by the flow of materials in forward direction and information and money in backward direction between the business entities. A supply chain is operated based on its existing business entities, and their facilities and networks. In the last decade, supply chain management has become animportant area due to an increasingly challenging marketplace. A particular problem with supply chains is known as demand amplification or Bullwhip effect in a supply chain occurring when slight to moderate demand uncertainties and variability become magnified when viewed through the eyes of management at each link in a supply chain (Russell and Taylor, 2003). A simple example of bullwhip effect canbe found in Kimbrough et al (2002). Bullwhip effect, also called Demand amplification, Whip-saw, Whip-lash effect, or Forrester Effect (Lee et al, 1979a), was first described by Forrester (1961) but the experts at Procter & Gamble gave its name. Lee et al (1979a & b) identified five major causes of the bullwhip effect, including demand signal processing, non-zero lead time, batched order,rationing game under shortage, and price fluctuations and promotions. Consequently, such effects influence costs, inventory, reliability and other important business processes in upstream supply chain members. The practical implication of this is the occurrence of various anomalous effects such as unusual stock levels, shortage of capacity for certain time periods, excessive variabilities in
20.1Proceedings of the Second International Intelligent Logistics Systems Conference 2006

labour requirements, late deliveries and obsolescence. This can result in severe problems for companies operating upstream in the chain, which are often mainly system induced. This leads companies to invest significant amounts in inventory management and scheduling systems only to see little improvement inperformance and profit. The only effective counter to the “Forrester effect” in supply chains is in the more effective communication flow, information sharing and integration along the supply chain. This in turn should result in a reduction of stock levels and lead times. The previous research in this area focuses on demonstrating the existence of this phenomenon and identifies its possible causes...