EURO 2024 Copenhagen
Abstract Submission

EURO-Online login

1357. Exact and non-exact approaches for the multi-stage stochastic two-dimensional cutting stock problem with usable leftovers

Invited abstract in session TA-29: Applications of combinatorial optimisation in industry and services I, stream Combinatorial Optimization.

Tuesday, 8:30-10:00
Room: 157 (building: 208)

Authors (first author is the speaker)

1. Adriana Cherri
Mathematics Department, UNESP, Bauru - INESC TEC, Portugal
2. Douglas Nogueira Nascimento
Institute of Mathematical and Computer Sciences, University of São Paulo
3. José Fernando Oliveira
INESC TEC, Faculty of Engineering, University of Porto
4. Beatriz Brito Oliveira
INESC TEC, Faculty of Engineering, University of Porto

Abstract

The Two-Dimensional Cutting Stock Problem with Usable Leftovers (2D-CSPUL) consists of cutting large plates to produce smaller ordered items, minimizing waste and allowing the generation and subsequent use of leftovers. This work explores a multi-stage stochastic approach to the 2D-CSPUL, incorporating demand uncertainty over multiple periods and enabling more adaptive and resilient production strategies. By adopting a scenario-tree approach, the study represents uncertainties throughout the planning horizon, offering a closer reflection of real-world scenarios. Additionally, anticipatory production of items ahead of immediate demand is allowed. Costs associated with the storage of leftovers and the inventory of items are also considered. The objective is to minimize material waste and storage costs of items and leftovers across all stages. The main contribution of this paper is a new mathematical model proposed to represent this problem. Due to its complexity, heuristic procedures based on the relax-and-fix strategy were proposed to achieve practical solutions efficiently. Computational results prove the effecitiveness of thre approach.
This research has been funded by the Fundação de Amparo a Pesquisa do Estado de São Paulo - FAPESP, Brazil (Process number 2022/05803-3 ) and Conselho Nacional de Desenvolvimento Científico e Tecnológico - CNPq, Brazil (Process number
317460/2021-8).

Keywords

Status: accepted


Back to the list of papers