Research On Inventory Risk Management And Control Technology Of Spare Parts Supply Chain Based On Cloud Platform

With the growth of China's economy and the improvement of people's requirements for travel convenience and comfort,domestic automobile production,sales and ownership increase year by year.The increase of car ownership also means that the market competition is more intense,the average age of cars is increasing.And customers are paying more and more attention to the quality of after-sales service.Therefore,the market demand for automobile manufacturers has shifted from satisfying the basic needs of customers and producing high-quality vehicles to improving the quality of after-sales service of automobiles.Differentiation and competition among enterprises are gradually inclined to the field of after-sales service.This change means that the automobile enterprises need to invest more human and financial resources in the construction of a more efficient and complete after-sales service system.In order to maintain the high quality of after-sales service,improve customer satisfaction and brand value.Vehicle enterprises need to ensure that after-sales service enterprises can get all kinds of vehicle accessories in time.And vehicle enterprises need to avoid service interruption due to disruption of parts supply.The factors that may lead to parts disruption are all kinds of unexpected risks on parts supply chain.In the thesis,a vehicle manufacturer(code name is CQ)is taken as the research object.And analyzed the supply mode,sales mode and inventory management mode of auto parts in detail.Also summarized potential risks on parts supply chain.Based on current research on inventory risk conduction model of supply chain,proposed that the system design and theoretical research can be carried out from inventory risk assessment,inventory risk prevention and inventory risk response in supply chain.Through the risk assessment of auto parts inventory,enterprises can perceive the risks existing in the current inventory structure.And take timely measures to prevent potential risks from turning into real risks.When inventory risk occurs,enterprises can use risk response mechanism to deal with unexpected risks efficiently and minimize losses.In view of above three perspectives,proposed using Bayesian network for inventory risk assessment.Using a new integrated model of dense adaptive cascade forest to classify accessories.And using iterative principal component analysis collaborative filtering algorithm for risk sharing collaborative enterprise screening.Because of the characteristics of Bayesian network theory,the application of Bayesian network in risk assessment can improve the interpretability of valuation results.Dense adaptive cascade forest is combined with integrated classification of deep forest and cascade boosting of boosting algorithm.These characteristics can make daforest improve classification accuracy layer by layer.Compared with random forest,SVM and MLP,daforest has higher classification accuracy.Iterative principal component analysis collaborative filtering can pad missing values and reduce data dimension with continuous iteration,and solve model scalability problem by clustering.Finally,based on the collaborative platform of automobile industry chain and.NET technology,the thesis realized the inventory risk management and the control system of auto parts supply chain.Through this system,enterprises can timely perceive inventory risk,formulate management strategies and quickly respond to unexpected auto parts supply risks,which reduces the management and maintenance costs of auto parts supply chain.