A MULTI OBJECTIVE MODEL FOR CONVERGENT SUPPLY CHAIN WITH CONTINUED PRODUCTION DESPITE THE FRACTION OF SOME PART

supply chains seek to reduce their costs and adopt a pure strategy to achieve competitive advantage. Companies therefore focus on the strategy of reducing inventories (raw material, product during construction, final product in stock) since inventory costs represent a significant part of total supply chain costs. But by adopting this strategy, supply chain versus Uncertainties and shortcomings are vulnerable. Therefore, companies should carefully balance the minimum costs and economic consequences of supply chain risks. One of the important decisions in utilizing the capacity of the assembly lines and preventing the stopping of production lines and sustaining the heavy costs of producing it, is the continued production despite the fraction of some parts. In this study, a three-echelon convergent supply chain, including a customer integrating other customers, a manufacturer, and several suppliers are considered. The customer will periodically announce to the manufacturer his approximated demand for the upcoming period. The manufacturer uses approximated demand and demand adjustment factor to calculate orders and determine the production plan. Also, in case of the possibility of assembling a product with fractional parts, instead of immediately transporting fractional parts, the uncompleted production policy is used and, upon the receipt of fractional parts, they are assembled on the final product. Therefore, reducing the premium freight, the cost of maintaining inventory, and the delay fine in delivery to the customer are conflicting goals, which requires multi-objective methods to solve. For this purpose, the problem is firstly modeled using robust optimization and the uncertainties in the problem are as far as possible addressed. Then, the results of the optimization are considered as the simulation optimization input involving all the uncertainties, then the output from optimization of the simulation is compared with the results of the robust optimization.