影响因子：

7.5

发表刊物：

Expert Systems with Applications

关键字：

The distributed manufacturing model enhances production flexibility and efficiency. This study investigates the order acceptance and scheduling problem in distributed heterogeneous factories while considering constraints including order-to-factory eligibility, processing time availability, and budget limitations, with the objective of maximizing total profit. The problem involves four interrelated decisions: order selection, allocation, sequencing, and scheduling. To address this, we formulate a Manne-based mixed-integer linear programming (MILP) model and a constraint programming (CP) model. However, these methods become computationally impractical for large instances. To improve efficiency, we develop three logic-based Benders decomposition (LBBD) algorithms that iteratively solve a master problem (MP) for order acceptance and allocation and subproblems (SPs) for scheduling within each factory. A local search (LS) algorithm is integrated to enhance solution quality and convergence speed. Experimental results validate the effectiveness of the proposed approach. For small-scale instances, our method achieves optimality where MILP and CP fail. In large-scale cases, it reduces the optimality gap by 54.2% and 100% across two scenarios while cutting solution time by 20.6%, 84.2%, and 98.3% in three scenarios. Applied to real-world scenarios, it increases total profit by 4.9%, confirming its practical value. These results demonstrate that the proposed LBBD framework significantly outperforms MILP and CP, solving larger instances efficiently with superior solution quality. Additionally, sensitivity analysis highlights key trade-offs in due date flexibility, budget constraints, and order distribution for optimizing scheduling performance.

合写作者：

景琳,向成飞,韩益彤

第一作者：

熊福力

论文类型：

期刊论文

论文编号：

127586

学科门类：

交叉学科

文献类型：

J

卷号：

280

是否译文：

否

发表时间：

2025-01-01

收录刊物：

SCI

发布期刊链接：

https://doi.org/10.1016/j.eswa.2025.127586