Abstract
Dynamic pricing problem contains the issue such a product price, storage capacity and so on, which influence the enterprise profit sorely. Thus our research will focus on the dynamic pricing problem to discuss. The object in this paper is to decide the product sales volume to maximize the enterprise profit. The past literature researched on the product with classless pricing problem. In this paper we will focus on the product such TFT-LCD and DRAM which can be classified into multiple quality grades (low-grade and high-grade). In this research we will use stochastic dynamic programming to derive and build the mathematical model, and then solve the model by applying the value iteration algorithm under the replenishment of inventory, independent and myopic environment. The result will be administered to the enterprise to make decisions on high-grade product sales volume, ordinary grade product sales volume, and high-grade product to make ordinary grade product sales volume at every sales period which can be maximize the enterprise total profit. This research also develops a dynamic selling decision-making system to evaluate the validity of the mathematical model. Through the comparison of present paper proposed the uncertainty pattern and the traditional pattern on the profit under the same assumptions, it may proof the product dynamic selling decision-making policy to be possible to promote the enterprise profit. Keywords: Dynamic selling decision-making policy、Dynamic pricing、Dynamic programming