Scheduling changes on the production floor are common in practice. These changes are aimed to meet the consumer demand, but they cause nervousness. Moreover, the nervousness will increase costs and reduce service level. This research deals with production batch size and buffer stock. It also considers to any changes in the preliminary order. Any changes in the demand for a given day are announced one day before and it happends randomly. This research considers two-echelon supply chain system with a single supplier and a single manufacturer. The models development are transactional relationship and consignment stock contract relationship. Additionally, this study also considers backorder and production capacity according to the real condition. Numerical examples are given to demonstrate the performance of the models. From the numerical results, it appears that coefficient variation (CV) of the demand affects the results obtained using method of Silver-Meal(SM) and Least Unit Cost (LUC).