Ordering Policy Under Continuous Demand Background With Bayesian Information Updating

With the rapid development of information technology, the consumer's demand are more and more diversified and individualized, the incertitude of market demand are growing serious, too. In the traditional supply chain's models, market demand is a random variable, and subject to a stable distribution. Now days the market environment is being more unpredictable, market demand will no longer be considered as a random which subject to a stable distribution, its distribution is also changing. Which has forced retailers to make quick response to the changing customer market, so they are increasingly concerned about the use of information on market demand in order to grasp the market better and enhance customer service level.The Bayesian method of demand information updating is adopted in this paper, we consider continuous demand background, single product, single manufacturer and single retailer supply chain system. From the perspective of the interests of retailers, we discuss a two-stage and multi-stage procurement strategy. In two-stage order problems, decentralized decision-making models and centralized decision-making models were established under assumptions that demand independent and demand related respectively. According to the different models, we give corresponding solving method, and then design a reasonable supply chain coordination mechanism. In multi-stage order problems, establish models to maximum the retailer's profit in every stage, combine with the signal of each stage and prior information, update demand distribution several times, then do a large number of simulation of the real market demand, analyze and compare the results.Numerical examples and sensitivity analysis of key parameters are given to all models in this paper. We compare the order policy and profit of models with Bayesian demand information updating and models without Bayesian demand information updating. We concluded that:Bayesian demand information updating method can eliminate the uncertainty of market demand to some extent, which help retailers to know the market demand better, adjust strategies to market demand changes in time and get better earnings at the same time. Besides, the profit of centralized decision-making is always higher than decentralized decision-making, but it's not that both two parties can benefit from the integrated supply chain model, a reasonable coordination mechanism could encourage both sides to accept cooperation and achieve win-win.