Research On Newsvendor Decision Problem With Incomplete Information

With the rapid development of market economy, the environment of competition the enterprise facing is more and more fierce. In order to occupy a good position in the fierce competitive environment, the enterprise is paying more and more attention on inventory management. Purchasing is the source of the inventory. Therefore, the quantity the enterprise purchases will directly effect the stock volume number. Newsvendor problem is a classical inventory model of operations management, and it is also a single period inventory problem for perishable products. It is to find the optimal purchasing quantity which maximizes the profit of the retailer in the uncertainty market. Information is important for retailer’s decision. However, due to the fast-changing markets and the accelerating of productions upgrading, and the independent decision-making of the participants in the supply chain making the asymmetric information between the enterprises, therefore, the retailer can not fully obtain the information of the market demand and her supplier’s cost structure when she makes decisions. The problems just mentioned lead to the loss of the retailer’s profit and the inefficient of the supply chain. Therefore, the newsvendor decision problem under incomplete information becomes a popular topic for studies.Because the retailer can not accurately know the demand distribution and the supplier’s cost structure, this paper studies the newsvendor decision problem when the retailer have partial information of demand with distribution-free approach, and studies the purchasing decision problem when the retailer have incomplete information of the supplier’s cost structure with incentive mechanism. So the research of this paper has important meaning in learning value and practice application. This paper mainly talk about following questions:Firstly, this paper investigates a joint pricing and purchasing problem for the newsvendor who distributes her products through the traditional retail and Internet at the same time with the assumption that only the mean and variance of the demand are known. Due to the assumption that the newsvendor only knows partial information about demand, this paper obtains the close-form expression for the robust optimal policy with distribution-free approach. Finally, we conduct numerical experiments to analysis the effects of the changes of the information and several parameters on the optimal policy and on newsvendor performance. Analysis shows that, the prices given by the distribution-free approach is less than the optimal risk-less prices. However, the profit under demand uncertainty is just slightly below the profit under demand certainty.Secondly, this paper studies the ordering decision of the retailer and the production decision of the supplier that the retailer have incomplete information about the market demand and there is random yield production. This paper examines the optimal decisions on ordering and production in a centralized, a decentralized and a emergence production cost sharing of random production supply chain. And we obtain the explicit expressions of the above three cases with distribution-free approach. Finally, we conduct numerical experiments to analysis the effects of the random yield, emergence production cost sharing and some important parameters on the optimal policy of the retailer and the supplier and on the performance of the supply chain system.Thirdly, this paper studies how the newsvendor uses the existing inventory with incomplete information about the demand to balance the product sales and project investment to optimize her target when she is capital constrained. We use the maximin rule of profits to deal with the case that the newsvendor only knows the mean and variance of demand while using the minimax rule of regrets to deal with the case that the newsvendor only knows the support of the demand, and we obtain the explicit expressions of the above two cases. Finally, we conduct numerical experiments to analysis the effects of the amount of information on the optimal policy and on the loss of performance. Analysis shows that the newsvendor will loss more if she has less information.Lastly, this paper studies the ordering decision of the retailer when she has incomplete information about the cost of the supplier. We discuss the problem that the disadvantaged retailer bargain with the advantaged supplier when the optimal ordering policy is not consistent with the optimal production decision. Two bargaining models are established for the supplier and the retailer under the condition of complete information and incomplete information. The optimal solutions of bargaining for the retailer in the above cases are attained. And according to that the least paying that the retailer should pay for the supplier and the information value of the retailer under the bargaining optimal solution are suggested. On the analysis of the values, the best decision of some important values for the retailer in complete information, the minimum cost and impacts of giving the supplier the minimum payments are discussed, as well as the impacts on the retailer’s optimal design plan in incomplete information. Analysis shows that the cost could be reduced for the disadvantaged retailer so as to achieve the coordination of supply chain by bargaining.