A learning-based solution method for practical slab allocation problem in multiple hot rolling lines

Ying Meng, Qingxin Guo, Qingyang Wang and Haichao Wang

IISE Transactions, 2024, vol. 56, issue 9, 988-1000

Abstract: Steel slabs are key in-process products in the steel production process. The slab allocation problem is to allocate slabs to suitable orders by considering complicated technical restrictions and multiple management requirements. In practice, thousands of slabs in multiple production lines should be allocated to orders. Due to the complexity and large scale of the problem, it takes planners a long time to make decisions, and the obtained allocation schemes are usually ineffective because of the myopic nature of the rule-based methods. In this article, we present a learning-based solution method for solving a practical slab allocation problem in multiple hot rolling lines. We formulate the problem as an integer programming model, then a data-based method is adopted to evaluate the mismatching cost between slabs and orders. To effectively solve the large-sized problem, a learning-based decomposition strategy is proposed to decompose the original problem into several small-sized subproblems. Then, a branch-and-price algorithm is proposed to optimally solve the subproblems. To further speed up the solution process, a primal heuristics based on column generation is designed to solve the large-sized subproblems. The solution methods have been implemented in a steel company and effectively increased slab utilization and reduced production cost.

Date: 2024
References: Add references at CitEc
Citations:

Downloads: (external link)
http://hdl.handle.net/10.1080/24725854.2023.2204316 (text/html)
Access to full text is restricted to subscribers.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:taf:uiiexx:v:56:y:2024:i:9:p:988-1000

Ordering information: This journal article can be ordered from
http://www.tandfonline.com/pricing/journal/uiie20

DOI: 10.1080/24725854.2023.2204316

Access Statistics for this article

IISE Transactions is currently edited by Jianjun Shi

More articles in IISE Transactions from Taylor & Francis Journals
Bibliographic data for series maintained by Chris Longhurst ().