The discrete parallel machine makespan scheduling location (ScheLoc) problem is an integrated combinatorial optimization problem that combines facility location and job scheduling. The problem consists in choosing the locations of $p$ machines among a finite set of candidates and scheduling a set of jobs on these machines, aiming to minimize the makespan. Depending on the machine location, the jobs may have different release dates, and thus the location decisions have a direct impact on the scheduling decisions. To solve the problem, it is proposed a new arc-flow formulation, a column generation and three heuristic procedures that are evaluated through extensive computational experiments. By embedding the proposed procedures into a framework algorithm, we are able to find proven optimal solutions for all benchmark instances from the related literature and to obtain small percentage gaps for a new set of challenging instances.
TS-EoH: An Edge Server Task Scheduling Algorithm Based on Evolution of Heuristic
School of Computer Science Central China Normal University Wuhan, China yatong\;School of Computer Science Central China Normal University Wuhan, China
Zuse Institute Berlin(柏林祖斯研究所);Hochschule für Technik und Wirtschaft Berlin(柏林工程与经济应用技术大学);Google Research(谷歌研究院);Technische Universität Berlin(柏林工业大学)
MAP-Elites based Hyper-Heuristic for the Resource Constrained Project Scheduling Problem
Commonwealth Scientific and Industrial Research Organisation(联邦科学与工业研究组织);Indian Institute of Technology(印度理工学院);University of New South Wales(新南威尔士大学)
Deep Reinforcement Learning Guided Improvement Heuristic for Job Shop Scheduling
Nanyang Technological University(南洋理工大学);Singapore Management University(新加坡管理大学);Institute of Marine Science and Technology, Shandong University(山东大学海洋科学与技术研究院);Eindhoven University of Technology(埃因霍温理工大学)
Guiding Large Language Models with Genetic Programming-Evolved Heuristic Knowledge for Dynamic Multi-Mode Project Scheduling
Centre for Data Science and Artificial Intelligence(数据科学与人工智能中心);School of Engineering and Computer Science(工程与计算机科学学院);Victoria University of Wellington(惠灵顿维多利亚大学)
Genetic Programming with Reinforcement Learning Trained Transformer for Real-World Dynamic Scheduling Problems
Digital Port Technologies Lab, School of Computer Science, University of Nottingham Ningbo China(诺丁汉大学宁波校区计算机科学学院数字港口技术实验室);School of Computer Science, University of Nottingham(诺丁汉大学计算机科学学院);Department of Engineering, University of Cambridge(剑桥大学工程学院);School of Engineering, Westlake University(西湖大学工程学院)
Low-Cost Labels, Reliable Choices: Rollout-Calibrated Hyper-Heuristics for Job Shop Scheduling
Faculty of Applied Sciences, Macao Polytechnic University(澳门理工学院应用科学学院);Pazhou Lab (Huangpu), Guangzhou(广州 Pazhou 实验室(黄埔));College of Animal Science and Technology, Zhongkai University of Agriculture and Engineering(仲恺农业工程学院动物科学与技术学院);Macao Polytechnic University(澳门理工学院)
Discovering Heuristics with Large Language Models (LLMs) for Mixed-Integer Programs: Single-Machine Scheduling
Grado Department of Industrial and Systems Engineering, Virginia Tech, Blacksburg, VA(弗吉尼亚理工大学工业与系统工程系);Department of Computer Science, Virginia Tech, Arlington, VA(弗吉尼亚理工大学计算机科学系)
Models and Strategies for Variants of the Job Shop Scheduling Problem
Cork Constraint Computation Centre(科克约束计算中心);University College Cork(科克大学学院);CNRS(法国国家科学研究中心);LAAS(法国航空航天实验室);Université de Toulouse(图卢兹大学);UPS(图卢兹第三大学);INSA(法国国立应用科学学院);INP(法国国立理工学院);ISAE(法国高等航空航天学院);UT1(图卢兹第一大学);UTM(图卢兹第二大学)
Learning Memory-Enhanced Improvement Heuristics for Flexible Job Shop Scheduling
Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education(教育部符号计算与知识工程重点实验室);College of Software(软件学院);College of Computer Science and Technology(计算机科学与技术学院);Singapore Management University(新加坡管理大学);Nanyang Technological University(南洋理工大学)
Tramp Ship Scheduling Problem with Berth Allocation Considerations and Time-dependent Constraints
Instituto Tecnológico y de Estudios Superiores de Monterrey(蒙特雷技术与高等教育学院);Pontificia Universidad Católica de Valparaíso(瓦尔帕莱索天主教大学);Universidad Autónoma de Nuevo León(新莱昂自治大学);Universidad de Los Andes(洛斯安第斯大学)
Accelerated Genetic Programming Hyper-Heuristics for Simulation-Based Scheduling via Agentic AI
Research Engineering Team REANNZ New Zealand;Centre for Data Science;Artificial Intelligence \& School of Engineering;Computer Science Victoria University of Wellington New Zealand
Bilevel Learning Model Towards Industrial Scheduling
Hong Kong University(香港大学);University College London(伦敦大学学院)
自动调度是制造业中广泛需要的一种调度方法,其目标是在有限的资源范围内优化作业顺序。然而,现有的调度系统严重依赖于启发式算法,这要么产生无效的解决方案或计算效率低下时,作业规模的增加。因此,开发新的大规模算法,不仅是高效和有效的,而且能够满足复杂的约束在实践中是非常重要的。在本文中,我们提出了一个双层深度强化学习模型,\textit{BDS},其中较高的水平是负责探索一个初始的全局序列,而较低的水平是旨在开发部分序列的改进,这两个层次是通过滑动窗口采样机制连接。在实现中,双深度Q网络(DDQN)用于上层,图指针网络(GPN)位于下层。在理论上保证了BDS的收敛性之后,我们在工业自动化仓库的场景中对其进行了评估,每条生产线上的作业数高达$5000$。结果表明,我们提出的BDS显著优于两种最常用的算法,三种强大的深度网络和另一种双层基线方法。特别是,与现实世界中最常用的基于贪婪的启发式算法(需要近1个小时)相比,我们的BDS在3个最大数据集上的最大完工时间分别减少了27.5%、28.6%和22.1%,计算时间小于200 s.