Journal des structures en acier et de la construction

Proposing a new model based on Bayesian Network as automated planner of pickup and delivery in inventory for steel industry

Abstract

Vahid Golmah*, Shahram Bpzorgnia and Mina Tashakori

Today, material pickup and delivery is the most important process for inventory planning of manufactures. Human operators usually schedule resources for pickup and delivery that it needs high cost and time and mistake decision cause to tiredness and pressure work. This problem is more acute for the steel industry. Therefore, using of an efficient expert system based on Artificial Intelligence (AI) could eliminate limitations of human planner that it has not applied for inventory planning in steel industry yet. In order to imbed a learnable model from decision patterns of human planners in steel industry, we propose an automated planner for pickup and delivery in raw/semi products of steelmaking based on Relief Bayesian Network (RBN). The proposed approach is applied for Mobarakeh steel company that results show the proposed approach decides as same as human planner for inventory.