Study On Several Kinds Of Vendor Managed Inventory Models

With the shortening of product life cycles, and the increasing request for customization of made-to-order products, demand forecasting has become one of the most difficult tasks for both retailers and suppliers in a supply chain. In addition, due to the Bullwhip Effect, the demand variation of the suppliers is much greater than the variation of the retailers. Thus, strengthening strategic partnerships between retailers and suppliers is critical to the further deveopment of supply chain management (SCM). Vendor managed inventory (VMI) is one of the SCM programs those can achieve smoothing information sharing and efficient Bullwhip Effect controlling. In a VMI supply chain, systematic and integrated inventory management is carried out, so that the whole supply chain is able to perform synchronically and efficiently.Based on the research of independent inventory and VMI, this paper studied the impact of VMI on the Bullwhip Effect, analyzed the parameters' sensitivity to VMI performance, in a model with auto regressive demand process AR(1), the demand between the retailers are correlated, also. Several kinds of VMI models were developed involved some factors those are more realistic, such as partial backlogging, detertioration, time-limited free back-orders, time value of money, and so on.On the basis of the VMI models without backlogging and with completely backlogging, new models for detertiorating items with partial backlogging and time-limited free back-orders were developed respectively, including a single vendor and multiple retailers, from raw material to finished prodect. Numeric examples and sensitivity analysises were presented to illustrate the models, and demonstrated that each parameter has different sensitivity to the costs of either VMI system.Taking account of the time value of money in the VMI model, the author studied the inventory replenisment decision with partial backlogging. Numeric examples and sensitivity analysises showed that the discount rate of money, service level and partial backlogging rate affect the costs of VMI system in their own ways. The theoretical evidences are provided for the VMI system to make better management decisions.