Demand satisfaction in supply chain management system using a full online optimal control method

In the business world, there are two key factors: customer satisfaction and cost. Delivery of goods from the manufacturer to customer is defined by supply chain management system systematically and the effect of the mentioned elements is seen in this structure. In this paper, current processes in a multi-product, multi-echelon supply chain are formulated as an updated discrete time dynamic model with work-in-process products. The model includes one raw material supplier, two plant (production unit) for manufacturing two types of products, one central warehouse, two local warehouses, and four retailers, and communication channels between the nodes are defined based on sharing possibility. Regarding advantages of decentralized control faced with multi-agent systems and functionality of predictive control to comply customer demand in a reasonable lead time according to predictive structure of supply chains and demands, a full online decentralized predictive control method with lot sizing is designed and applied to the model. The simulations show a trade-off between reduced cost based on lot sizing and customer satisfaction obviously.