供應中斷是OEM供應鏈中企業面臨的主要風險。本文基于供應鏈彈性分析的角度,將OEM供應鏈彈性運作問題描述為多變量耦合控制模型,構建了可變結構的彈性控制系統,研究了在供應中斷風險沖擊下OEM供應鏈彈性交互影響機制。在此基礎上,提出了一種有針對性的提升供應鏈彈性的深度學習機制,此算法比傳統的BP神經網絡更加能夠提高供應鏈績效,并結合案例進行驗證。研究結果表明:當供應中斷發生時,深度學習算法可有效提升OEM供應鏈彈性,最大程度減輕企業損失。

Because the OEM supply chain may face greater disruption risk than ordinary supply chain,in this paper,the elasticity operation and promotion strategy for OEM supply chain are mainly studied.OEM supply chain resilient operation problem is described as a multivariable coupling control model,constructing the resilient control system of variable structure,researching the impact of supply chain resilient interaction mechanism undersupply disruption risk.On this basis,a kind of deep learning mechanism is put forward to improve the flexibility of OEM supply chain.This algorithm can improve the performance of OEM supply chain more than the traditional BP neural network.The results show that:when the supply disruption occurs,the deep learning algorithm can effectively enhance the OEM supply chain flexibility,and it can reduce the pecuniary loss of the enterprise to the maximum extent.