A simulation-based optimization framework for parameter optimization of supply-chain networks

This work presents a novel approach that addresses the management of chemical supply chains (SCs) under demand uncertainty. One of the main objectives is to overcome the numerical difficulties associated with solving the underlying large-scale mixed integer nonlinear problem (MINLP). The approach that is proposed relies on a simulation-based optimization strategy that uses a discrete-event system to model the SC. Within this framework, each SC entity is represented as an agent whose activity is described by a collection of states and transitions. The overall system is coupled with an optimization algorithm that is designed to improve its operation. This strategy is a very attractive alternative in the field of decision-making processes under uncertainty, the advantages of which are highlighted with some cases of SC networks that are composed of several plants, warehouses, distribution centers, and retailers.