Share
Export Citation
Control of Inventory Systems in Probabilistic Environment via Chance Constrained-Based Robust Predictive Controller
Widowati
International Review of Automatic Control
Q3Abstract
This study has been used to solve the control problems of inventory systems in a probabilistic environment. This has involved the use of an inventory system or warehouse with one type of product purchased from a supplier with some time needed for transportation and some of the products considered to be demanded by buyers or consumers. Meanwhile, some have been considered to be defected before entering the warehouse due to several reasons such as damages during delivery or bad quality. This has led to the use of the demand value and defect product rate as the two probabilistic parameters, and they have been treated as random variables with some probability density functions. Moreover, the inventory level behavior represented by the number of products stored in the warehouse has been modeled as a linear dynamical system containing the two aforementioned probabilistic parameters. The manager of the warehouse has been also considered the one deciding the number of items to be purchased from the supplier at every period in order to ensure the expected demand is satisfied and the inventory level is close to a “safe” point. Furthermore, a robust model predictive control, which is a model-based control method, has been later applied to this inventory control problem. The numerical experiment results have showed that the optimal decisions have been achieved and the inventory level has been sufficiently closed to the “safe” level. Therefore, the approach proposed in this study is recommended to be adopted by managers of inventory systems.