Research On Bi-Level And Multi-Objective Conditional Value-at-Risk Model

In recent years, with the development of economy and society, more and moreuncertain factors are changing rapidly, and the supply chain system is facing all sortsof risk impact. Generally, supply chain system structure is complex, and risk cancause great and irreparable loss. The traditional opinion of pursuiting maximizedprofits barely is not working, and more and more enterprises consider the control ofrisk as an important basis for decision-making. Effective risk management can notonly control the risk, minimize the suffer of the possible loss, but also improve thesupply chain, reduce the cost, make the enterprises in the supply chain easily tosurvive and develop. Establishment of bi-level and multi-objective programmingmodel can make a more comprehensive study of supply chain risk problem, and usethe scientific quantitative methods to improve the capacity of the risk analysis and riskmanagement.In this paper, we use the Conditional Value-at-Risk theory, bi-level programming,multi-objective programming, penalty function theory, supply chain management, riskmanagement and other fields of knowledge, to establish the bi-level andmulti-objective Conditional Value-at-Risk model.We give the the equivalent modelwhich is easy to be solved, and complete the proof of equivalence. Using K-Ttheorem, convert it into one-level programming model.Then, we give the penaltyfunction algorithm, to get the approximate optimal solution by the Matlab and otherrelated software. On the basis of theoretical research, we apply the general model tothe supply chain management practice, build bi-level and multi-objective ConditionalValue-at-Risk models about one-supplier and one-retailer, do numerical analysis,study the risk mechanism, and prove the validity of the model. In this paper, we establishes the general model can be widely used in the supplychain management practice, set various specific models of single product and multipleproducts about procurement, production, inventory, transportation and marketing, tohelp the enterprises to avoid risk and make more effective management decisions.