| Inventory management models are widely used in practical world and they can help the decision-makers to obtain the optimal production or ordering lot sizing effectively. For the traditional inventory management models, researchers always assume 100% of items in an ordered lot are perfect and the demands is constant value, then make decision to the production or ordering lot sizing. But in fact, due to imperfect production process, natural disasters, damage or breakage in transit, or many other reasons, the lot sizes received may contain some defective items. In addition, the demand and the rate of the defective products can be influenced by many factors, so they are nondeterministic. Moreover, with the development of supply chain management, the research of inventory management should not be restricted in one enterprise. So, when we design a complex inventory system, these factors should be considered comprehensively.First, the paper researches retailer's order model in deterministic environment. Combined the characteristics that customers of retailer don't accept the defective products and vender promise that defective products can be send back. Considering the order lot and the order frequency of every return cycle are variable, build the economic order quantity model which maximum expected profit per unit time. Then, through the comparison with a classic EOQ model example, demonstrate the feasibility of the order model in such situation.Then, based on the study in deterministic environment and fuzzy credibility theory, considering the demand and the rate of detective products are fuzzy, take (R,Q) inventory control strategy for example, build the inventory model to satisfy the service level destined. Obtain the inventory control strategy in fuzzy environment and analyze the influence of the fluctuation of demand and the rate of defective products on quantity of order, the order frequency of return cycle, safety stock and reorder point. At last, the paper researches the inventory control strategy in supply chain environment. Considering the demand and the rate of defective products are fuzzy, based on principal-agent theory and common replenishment epoch (CRE), build a contact which take minimize the cost of both vendor and retailer as a target. At last, an example is given to demonstrate the feasibility and superiority of the contact. |