Reliability Analysis And Pattern Recognition Of Supply Chain System In Manufacturing Industry

In May of last year,BOSCH group failed to provide steering gear to BMW timely,which made the shutdown of Tiexi plant.In operation of supply chain system,there are many uncertain factors,which often bring inestimable loss to the system and even make the system disrupted.If we can effectively grasp and measure the reliability of the supply chain system,it will promote the healthy and stable development of the system and all enterprises.Based on the above background,in order to help enterprise managers and supply chain managers to capture supply chain's reliability,we use the artificial neural network intelligent algorithm and reliability theory to establish a four level supply chain model which is based on manufacturing enterprise.This thesis studies the reliability of manufacturing industry supply chain from the whole point of view.The factors affecting the reliability of supply chain system are determined by the quality control tool 5M1 E method(man,machine,material,method,environment and measurement),They are people,node enterprises,materials,funds,information,environment and cooperative strategies.In order to grasp the reliability of supply chain system quantitatively and accurately,we set up the inventory of manufacturing enterprises on the supply chain.Based on two cases:first,there is no failure and idle of all manufacturing enterprise,second,there is only one manufacturing enterprise failure and its inventory is not empty.We adopt Mark-off dynamic transfer method to quantitatively study the reliability of the manufacturing enterprise and its inventory,and the calculation formula of the steady-state availability of the four level series supply chain system is derived.Then,the system is divided into three subsystems: production,supply and marketing,and we use fault diagnosis to study the supply chain reversely.The fault tree analysis(FTA)is used to establish a fault tree with the failure of the manufacturing supply chain system as the top event.The most critical fault factor is determined by the minimum cut set.In order to make the model more close to the actual situation,the BP neural network algorithm is applied to simulate the supply chain status.The reliability state of the system is divided into five categories: healthy,stable,good,poor and unreliable.The BP neural network is used to simulate the operation state of the system,and the effectiveness of the BP neural network is verified by a practical example.