A Lagrangian relaxation for flexible order batching problem in iron and steel industry

This paper investigates the flexible order batching problem in iron and steel industry. The problem consists of grouping a set of orders to form a set of production batches - charge-lots. A novel mixed-integer programming model is established by taking the flexibilities of orders and the technological limitations into consideration. A solution methodology is developed combining Lagrangian relaxation, dynamic programming, and heuristics. Computational results using practical data collected from an iron and steel industry and random data of large scale problems, demonstrate that the proposed method generates high quality solution within a reasonable computation time.