Research On Decision Of Server Purchase Order Allocation In K Company Under Uncertain Supply Environment

As the Internet application continues to heat up,Internet traffic and scale of users has increased significantly.It is expected that Internet traffic and scale of users will continue to increase for a long time in the future,and the server is the supporting foundation for Internet companies' business operations,so the demand for servers will continue to increase in the future.However,the current uncertainty in the supply and demand of the server market has also led to the instability of the server procurement environment.For suppliers and buyers,it is a challenge to ensure the effective supply of servers.The multi-vendor procurement strategy adopted by K company can reduce the supply risk to a certain extent,but K company did not consider the server supply uncertainty in the process of implementing purchase order allocation decisions,resulting in K company's procurement costs have increased significantly and the number of servers purchased may not be able to effectively meet demand,meanwhile the suppliers' supply pressure is high.Based on the above background,this paper studies the problem of server purchase order allocation of K company's under uncertain supply environment.(1)This paper analyzes the current status of server purchase order allocation decision of K company.First,it provides an overview of K company's business and the demand of servers.At the same time,it analyzes the uncertainty of K company's server supply and the impact of this uncertainty.And then it analyzes the problems of current K company's server purchase order allocation method.Finally,in view of the existing problems of server purchase order allocation,a decision model for K company's server purchase order allocation considering the uncertainty of supply is established and solved.(2)This paper uses literature analysis to understand the current research status of purchase order allocation decision-making.Based on the existing research results,the model analysis method is used to establish a two-stage multi-source procurement integer programming model that considers quantitative flexible contracts.The model is a multi-objective problem model and this paper selects the NSGA-II algorithm as the main algorithm for the solution.Using the on-site survey method,this paper conducts a field survey of the actual procurement activities of K company,and substitute its actual supply and demand data into the model for solution and analysis.(3)According to the actual demand and supply data of K company,Matlab was used to implement the NSGA-II algorithm,and the decision-making plan under different preferences of K company under different random supply scenarios was obtained.In this paper,the solution obtained by solving the model optimizes the shortcomings of the original decision method,and the two objective function values obtained are also better than the two objective function values of original plan.It is obtained that the two different preference decision-making schemes obtained by the model are sensitive to the change of the demand quantity.The decision plan that prefers the ! goal is not sensitive to changes in unit out-of-stock costs.The model is suitable for the server purchase order allocation problem of K company.In general,when the demand changes,the diversity of the optimal solutions obtained by the model is good.When the unit out-of-stock cost changes,the stability of the model is good.K company should pay attention to the changes in demand to ensure the diversity of decision-making schemes,and at the same time,in the future development process,pay attention to the value changes of users and reduce the loss caused by out of stock.