Developing an Optimization Algorithm for Multi-product and Multi-level Inventory Systems with Random Parameters
Subject Areas : Industrial Management
Fariborz
Jolai
1
(Professor, Department of Industrial Engineering, University of Tehran, Tehran, Iran)
Sayyed Mohammad Reza
Davoodi
2
(Ph.D. Student, Department of Industrial Management, Dehaghan Branch, Islamic Azad University, Iran,)
Ali
Mohaghar
3
(Associate Professor,Department of Industrial Management, University of Tehran, Tehran, Iran)
Mohammad Reza
Mehregan
4
(Professor, Department of Industrial Management, University of Tehran, Tehran, Iran)
Keywords:
Abstract :
The present study aimed to develop and compare a simulated model of multi-product and multi-level inventory systems. The model is developed for the final product, different medium products, and the main product. The main purpose of optimization is to minimize costs function. The servicing level of units is measured through backfilling rate that should be more than a minimum level. In the proposed algorithm, the local optimization was found through the genetic algorithm. Since point estimation of goal function and backfilling rates are done on the simulation in the present study based, the statistical methods were used for investing possibility of solutions. Finally, one example was presented in the three-level network. Because linear localization is an especial form of second-order localization, the difference between goal function and estimated volume was at the minimum level. Undoubtedly, it is expected that the estimated point of this algorithm is better than the estimated point of linear localization.
_||_