The Mechanism Of Risk Transmission Related To Supply Chain Disruptions And Risk Predictions Of Supply Chain Disruption

In recent years, with the advance of business modes such as global purchase in enterprises, non-core business outsourcing, single-source supply, agile production and lean supply, supply chains expand in space and shorten in time. The external factors such as natural disasters, economic fluctuations, epidemic diseases, terrorism, and wars are likely to occur. The changes in space and the external disturbance in supply chains accelerate the uncertainty and fragility of operational environment of supply chains and increase the possibility of disruption occurred. The disruption of node enterprises in supply chains may transmit in the system of supply chains with the butterfly effect, and the growing effect of disruption transmission can damage the safe operation of supply chains. Eventually, disruption, invalidity, paralysis, and collapse may emerge in the overall supply chains. Various events of disruptions in supply chains happen in succession, affecting the smooth and steady production of enterprises, resulting in a tremendous amount of loss in assets for enterprises.With respect to the complexity of risk transmission of supply chain disruptions and the uncertainty of demand, this study involves the mechanism of risk transmission related to supply chain disruptions, develops and solves the problem of prediction after the risk transmission of disruptions, which is an important concern in supply chain management. The primary object of this project is the risk of supply chain disruptions. The endogenous and exogenous risks in supply chain disruptions are viewed as decision-making variables, and their property, connotation, characteristics, category, factors, and internal mechanism are systematically investigated, and the features of risk transmission in supply chain disruptions are considered in different network structures. In the environment of complexity, uncertainty and fragility of network operations in supply chains, the product demand is highly unpredictable, and the research on prediction of supply chain disruption so far has not provide a sound solution. For this reason, this study aims to build a predicti on model of risks in supply chain disruptions, and the empirical analyses are also presented. Four major findings are of great importance below.Firstly, this study analyzes the concept, characteristics and causes of transmission in relation to risk transmission of supply chain disruptions. It discusses the risk transmission elements in supply chain disruptions in detail and depicts the graph of their logic relations. It expounds the risk transmission path of supply chain disruptions and draws its transmission chart. The risk transmission process can be divided into five periods: latent period, incubation period, outbreak period, involute period, and reservation period. It considers the complexity of risk transmission of supply chain disruptions from the perspectives of the complexity of the structure of supply chain networks, transmission paths, and the dynamic sophistication of transmission processes. These analyses and discussions are conducive to revealing the intrinsic mechanism of risk transmission of supply chain disruptions and the transmission pattern of disruption risks. Also,they contribute to a better understanding of the structural stability, coordination and optimization of each internal node enterprise in supply chains.Secondly, with regard to the complexity of risk transmission paths and processes in relation of supply chain disruptions, the small world network and its research method, and such statistic parameters as application, uniformity, length of characteristic paths, clustering coeffici ents, vertex degree, degree distribution, entropy of network structure, and alternating current frequency are introduced. It analyzes the properties of risk transmission of supply chain disruptions. It has been shown that there is a negative correlation between the risk transmission speed of supply chain disruptions and the transmission path and node numbers of the network; shorter risk transmission paths of disruptions and greater clustering coefficient are beneficial for sharing the information of disruption risks in supply chains. With core node enterprises, their stability plays a pivotal role in stabilizing the overall supply chain network in the risk transmission of supply chain disruptions. In the structure of small world network, the requirements of statistic parameters vary in different periods of risk transmission of supply chain disruptions.Thirdly, in consideration of variability of the scale of supply chain network in reality, the BA scale-free network is deployed to analyze the network features of supply chains in relation to the sophistication of risk transmission paths of the supply chain network and their transmission processes. It has been shown that the diversified weak connections in the structure of scale-free networks enable each node enterprise to build more risk transmission channels of disruptions compared with the structure of small world network; more new information of disruption risks with low redundancy can be obtained; disruption risks of random failures can be better averted, transferred, dispersed and handled; more and shorter risk transmission paths of supply chain networks can be built; time lag of information transmission of disruption risks between node enterprises can be shortened; the dependence of relaying transition of core node enterprises can be reduced; the risk of ossification of network structure of supply chains can be lowered.Finally, in the process of risk transmission, with high uncertainty of product demand, there is not any sound solution to the problem of predicting supply chain disruption temporarily. This study builds a model of risk predictions of sup ply chain disruption and then the empirical analyses are provided. The prediction model of improved grey neural network and the results of its empirical analyses show that the determination of the number of neurons in the input level of BP neural network through the optimal dimensions in the improved prediction model can break through the limits of the traditional grey model GM(1,1) on the initial data sequence, and it also reveals the good feature of nonlinear approximation of neural network. This increases regression level and the precision of prediction, enhances the stability and reliability of prediction, and resolves the predicting problem of supply chains soundly.