Research on mining collaborative behaviour patterns of dynamic supply chain network from the perspective of big data

Abstract

In view of the limitations of the linear model used in the traditional supply chain solution method, it cannot adapt to the dynamic supply chain network solution method in the multi-source big data environment of the Internet, and maps the dynamic supply chain into a network diagram model, and proposes e-commerce. Supply chain network model, based on this model, presents a semi-instance pattern detection method based on collaborative matrix decomposition, which is used to detect a semi-instantiated collaborative behaviour pattern in the supply chain network. According to the given collaborative behaviour model, the collaborative supply matrix decomposition method is first used to calculate the candidate supply chain of the personalized supply chain, and the degree of intimacy between node entities in the supply chain network is calculated. Using the A* graph search algorithm, a supply chain result candidate chain set is generated based on a personalized supply chain candidate set. According to personalized time, cost and other constraints, the final supply chain solution is tailored. The correctness, efficiency and accuracy of the method were verified by the e-commerce supply chain data set for apparel.