Research On Supplier Selection And Dynamic Order Allocation Problem Under Uncertainty

With the increasing production competition and the increasing requirements of users for product quality,reducing production costs,improving product quality,and increasing customer satisfaction are critical to the operation and development of enterprises.Optimal supplier selection and rational order allocation can help companies reduce the cost of the final product and improve quality,so it has received widespread attention.Supplier selection and order allocation problem involves typical two-stage decisions.When making the first-stage strategic supplier selection decision,many key information(demand,product quality,and so on)are uncertain.But we need to select the suppliers and sign contract with the selected suppliers before we have the accurate information.Then,according to the determined supplier selection decision and newly observed information,the second-stage operational order allocation decision will be made.Since the first-stage decision is made under uncertain information and will not change for a long time in the future,it is related to risk.Considering this,we conduct this research on the supplier selection and dynamic order allocation problem under uncertainty.The main contents of this study include the following two aspects.First,the two-stage decision-making for supplier selection and order allocation based on risk-mean analysis.For the supplier selection and order allocation problem with the uncertainty of future demands and operation costs,two decision criteria,value at risk(Va R)and expected value(EV),are introduced to evaluate the performance of the supplier selection decision in cost,then a fuzzy two-stage optimization model is constructed.And a new solution methodology to the proposed model involving fuzzy variables is inherently designed based on theoretical analysis and proof.Second,the two-objective decision-making for supplier selection and order allocation based on EV and conditional value at risk(CVa R).Since the supplier selection is a typical multi-attribute decision-making problem,in addition to the cost,the purchase value of the decision scheme is also considered.Thus the stochastic two-stage two-objective programming models based on the CVa R and EV criteria are proposed.Meanwhile,for solving the proposed models,we develop a hybrid solution algorithm combining heuristic algorithm with stochastic simulation and accurate solution.And extensive numerical experiments are used to illustrate the effectiveness of proposed models and algorithm.Different from the traditional research with two-stage idea that does not consider the effect of order allocation decision on supplier selection decision,the two-stage models constructed in this study fully consider the sequential effect between supplier selection and order allocation,which provides novel ideas for the research on supplier selection and dynamic order allocation under uncertain environment and helps to improve related theories.Besides,in order to evaluate the risk of decision brought by uncertain information,the performance of supplier selection decision is determined by using metrics such as Va R,CVa R and EV.Meanwhile,the analysis of results under different measurement provides more flexible schemes for the decision maker,which has practical guiding significance.