Robust simulation-optimization of dynamic-stochastic production/inventory control system under uncertainty using computational intelligence

Abstract

In production/inventory control systems, the goal of the controller is to generate sophisticated decisions by controlling the order rate and inventory level. This paper aims at modeling a dynamic-stochastic production/inventory control system under two sources of variability (uncertainty) including uncertainties on demand rate and frustrating rate. The study deals with obtaining a robust optimal design of a Proportional-Integral-Derivative (PID) controller in in the stochastic control system. For this purpose, a new robust simulation-optimization method in the class of computational intelligence is proposed. To cope with the unknown distribution of uncertainty, the crossing weighted uncertainty scenarios are combined with the proposed method. Within this study, a new sequential robust efficient global optimization is proposed to make a trade-off between optimal and robustness terms in final optimization results. Finally, a numerical case with simulation experiments is conducted to demonstrate the advantages of the proposed policy in terms of optimal result, robustness, and computational cost.