Designing a resilient supply chain through a robust adaptive model predictive control policy under perishable goods and uncertain forecast information

We deal with the inventory level control problem for Supply Chains (SCs) whose dynamics is affected by two sources of uncertainties: 1) perishable goods with uncertain deterioration rate, 2) an uncertain future customer demand freely varying inside a given bounded set. The purpose of our contribution is to propose a smooth Replenishment Policy (RP) maximizing the satisfied customer demand and minimizing the inventory level. These requirements should be satisfied despite the above uncertainties and unforeseen customer demand patterns trespassing the "a priori" assumed boundaries. To this purpose we define a Resilient RP (RRP) using a new Robust Adaptive Model Predictive Control (RAMPC) approach. This requires solving a Minimax Constrained Optimization Problem (MCOP). To reduce the complexity of the solving algorithm, we parametrize the predicted replenishment orders in terms of polynomial B-spline basis functions.