Supply Chain Coordination Based On Bayesian Demand Forecast Updating And Cvar Models

With the continuous progress of technology and the rapid growth of economic, market competition has become more and more fierce, making increasingly diversified products range and better products quality, and the identifying ability of customers are growing rapidly. Therefore, the demand uncertainty increasing highly--not only the type of demand distribution, but also the demand distribution parameters. How to control the risk and reduce demand uncertainty has been the big problem which should be solved firstly. The conditional value-at-risk (CVaR) as a risk measurement tool, because of its mathematical superiority, is widely used in various fields, including the supply chain. Furthermore, the Bayesian demand forecast updating is an effective method to reduce demand uncertainty by the fusion of the information collected and the prior information to update the existing information and amend the distribution of demand. These will be applied to the paper.This paper explores the ordering and supply chain coordination problems of the seasonal products which have a longer lead time and shorter sales time of two-stage supply chain which consists of a retailer and a supplier in a single cycle. This paper will take the lead time as a decision variable into the models, considering the work fees brought up by shortening the lead time. From two dimensions--decentralized decision-making model or centralized decision-making model, risk neutral or risk averse--established four models. When the decisioners are risk neutral, taking the maximized expected profits as the objective function, and when the decisioners are risk averse, taking the conditional value-at-risk (CVaR) as the objective function, then solute the models. In order to verify the validity of the model and algorithm, numerical experiments are present to compare and analysis the decision results from the models.Finally, consider the supply chain coordination problems. Designing contracts to balance the revenue:risk-sharing contracts and buy-back contracts:the work fees brought up by shortening the lead time bear by both sides with a certain proportion; at the end of sales season, the supplier buys back the unsold products with a certain proportion so that reduce the concern of retailer and ensure that the interests of retailer will not be less than the benefits under decentralized decision-making, ensuring the supply chain coordination. The validity of contracts was verified by numerical experiments and it also analyzes the impact of the lost sale penalty cost to the distribution of profits.