Risk Warning Of Supply Chain Based On Fuzzy Reasoning

Along with the trend of balancing the development between regions and coordinating the specialization of industry, the improving of socialization and specialization of production, modern logistics, as a foundational services industry, is becoming increasingly important in our national economic system.Modern logistics involves extensive domains, obtains much employment numbers, make a huge difference in promoting production and consumption, plays a significant role in industrial restructuring, accelerating transformation of the mode of economic development and enhancing national economic competitiveness. A survey conducted by Todd McGrath reveals that 10% total cost is saved, improves n-time delivery rate 15%, and makes Order - production cycle shortened about 25%-35%, helps the member of the supply chain enhancing the productivity approximately10%, and the most ,ameliorates the rate of fixed assets to 15%-20%.And today, accompanied the global economic integration, the enterprise is undertaking massive pressures, such as market demand with rapid changeability , cost reduction and high-quality follow-up services, all those makes it harder to delivery what the consumer needed to the right place in correct amount with good quality and proper state. Once some risks break out in one procedure and it will amplify many times in the transferring to other enterprise, even makes the supply chain collapse totally, resulting in inestimable economic losses. What needed is adopt advanced and precautionary means in risk management of supply chain, monitoring and warning the operation situations timely and effectively, coordinating the members to avoid blindness of production, ensuring the supply chain runs of health and sustainability.This paper builds the mechanism of risk early warning of supply chain in combining neural networks and fuzzy control theory. Trying to find out the relationship between each risk contributing factors by training the primitive data using neural networks, and then generate the fuzzy rules. By doing so, avoids the error caused by the improper understanding of fuzzy control rules designers and operators, guarantees the optimal or suboptimal performance of the fuzzy controller; Fuzzy reasoning rules are distributed across the network with whose ability of parallel high-speed distribution calculation, comes out the result of risk evaluation. With the result, factors may contributing to risks can be identified, then monitored and esteemed, the negative trend of supply chain system will be controlled and corrected.