Robust Optimization of Multi-product and Multi-class Lot-sizing and Supplier Selection with Uncertain Demand

In this study, a multi-product and multi-period lot-sizing and supplier selection problem has been considered. The demand of products is multi-class and uncertain. Due to the interchangeability of products, it is possible to satisfy some part of their demand with the alternatives. In the case of an inventory shortage, a specific part of the shortage will be lost sales and the remainder will be backorder. Suppliers can have an all-units quantity discount policy. For confronting uncertain demand, possible modes are defined as scenarios and the robust optimization approach proposed by Mulvey et al. is applied. The objective function of the problem, which is to be minimized, is made up of the total cost of purchasing, transportation, inventory, demand substitution, lost sales, and backorder. The upper and lower bounds for the number of suppliers per product family are defined as a constraint for the implementation of management policies for supplier selection. Dynamic and multi-class demand in a multi-period horizon, together with allowed backlog and lost sales, are features that, to the best of our knowledge, are not yet considered by other researchers. There are many supply chains that have sold the products and should supply required spare parts. The results of this research help attain optimal ordering of the parts. The performance of the model is examined with a numerical example.