Article
Engineering, Multidisciplinary
Tao Zhou, Liang Luo, Shengchen Ji, Yuanxin He
Summary: This study proposes a new method to solve the permutation flow shop scheduling problem (PFSP) using an end-to-end deep reinforcement learning approach to minimize the maximum completion time. Experimental results demonstrate the superiority of the proposed method in multiple metrics.
Article
Computer Science, Artificial Intelligence
Yu Du, Junqing Li, Chengdong Li, Peiyong Duan
Summary: In this study, a DQN model is proposed to solve a multiobjective FJSP with crane transportation and setup times. The model optimizes makespan and total energy consumption simultaneously based on weighting approach. The DQN model uses 12 state features and seven actions to describe the scheduling process, and applies a novel structure in the DQN topology. Extensive computational tests and comparisons demonstrate the effectiveness and advantages of the proposed method in solving FJSP-CS.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Engineering, Electrical & Electronic
Qi Yan, Wenbin Wu, Hongfeng Wang
Summary: This paper investigates the integrated optimization of distributed permutation flow shop scheduling problem (PFSSP) with flexible preventive maintenance (PM). It proposes a deep Q network-based solution framework, which exhibits strong solution performance and generalization abilities according to numerical studies, and also obtains a suitable maintenance interval and managerial insights.
Article
Computer Science, Interdisciplinary Applications
Jiang-Ping Huang, Liang Gao, Xin-Yu Li, Chun-Jiang Zhang
Summary: This paper studies the Distributed Job-shop Scheduling Problem (DJSP) with random job arrivals and explores a multi-agent method based on Deep Reinforcement Learning (DRL). The effectiveness of the proposed method is proven through independent utility tests and comparison tests, and its practical value in actual production is demonstrated through a case study.
COMPUTERS & INDUSTRIAL ENGINEERING
(2023)
Article
Automation & Control Systems
Ruiqi Chen, Wenxin Li, Hongbing Yang
Summary: In this article, a novel deep reinforcement learning framework is proposed for solving the classical job-shop scheduling problem. This method utilizes the attention mechanism and disjunctive graph embedding to model the problem and an improved transformer architecture to generate solutions. Experimental results confirmed the effectiveness of this approach.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Chemistry, Multidisciplinary
Christos Gogos
Summary: This paper investigates the permutation flow-shop scheduling problem and its distributed version, proposing constraint programming models and a novel heuristic to solve them. Experimental results demonstrate the effectiveness of the approach and highlight the significance of the number of jobs in problem complexity.
APPLIED SCIENCES-BASEL
(2023)
Article
Computer Science, Artificial Intelligence
Chupeng Su, Cong Zhang, Dan Xia, Baoan Han, Chuang Wang, Gang Chen, Longhan Xie
Summary: This paper proposes a framework based on Graph Neural Network and deep reinforcement learning to solve the dynamic job shop scheduling problem with machine breakdown and stochastic processing time. The model effectively extracts the embeddings of the state by considering the features of the dynamic events and the stochasticity of the problem. Evolution strategies are utilized to find optimal policies that are more stable and robust than conventional deep reinforcement learning algorithms. Extensive experiments demonstrate the superiority of the proposed method over existing reinforcement learning-based and traditional methods on multiple classic benchmarks.
APPLIED SOFT COMPUTING
(2023)
Article
Chemistry, Multidisciplinary
Chang-Bae Gil, Jee-Hyong Lee
Summary: This study solves the material scheduling problem of multiple machines in a hybrid flow-shop environment using deep reinforcement learning. It considers practical factors and proposes a method to simplify the high-dimensional environmental space for efficient learning.
APPLIED SCIENCES-BASEL
(2022)
Article
Engineering, Industrial
Lu Zhang, Yi Feng, Qinge Xiao, Yunlang Xu, Di Li, Dongsheng Yang, Zhile Yang
Summary: This paper investigates the difficulties of Dynamic Flexible Job Shop Scheduling (DFJSP) caused by the uncertainties and complexity in the production process due to customized requirements. A new DFJSP model, VPT-FJSP, is proposed and solved using Markov decision process (MDP) and reinforcement learning methods. The experimental results show that the proposed framework outperforms genetic algorithm and ant colony optimization in most cases, demonstrating its effectiveness and robustness.
JOURNAL OF MANUFACTURING SYSTEMS
(2023)
Article
Management
Janis Brammer, Bernhard Lutz, Dirk Neumann
Summary: This study presents a novel reinforcement learning approach for the permutation flow shop problem (PFSP) with multiple lines and demand plans. The approach generates job sequences iteratively and optimizes them using local search, outperforming existing methods on multi-line problems with short cutoff times.
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH
(2022)
Article
Engineering, Manufacturing
J. F. Ren, C. M. Ye, Y. Li
Summary: This study proposes a Nash Q-learning algorithm for Distributed Flow-shop Scheduling Problem based on Mean Field approach, which uses multi-agent reinforcement learning framework. By designing a two-layer online learning mode and approximating the value function with neural network, the proposed algorithm demonstrates feasibility and efficiency in solving large-scale production scheduling problems.
ADVANCES IN PRODUCTION ENGINEERING & MANAGEMENT
(2021)
Article
Computer Science, Artificial Intelligence
Minghai Yuan, Liang Zheng, Hanyu Huang, Kaiwen Zhou, Fengque Pei, Wenbin Gu
Summary: In this paper, an improved double Deep Q Network (DDQN) real-time scheduling method is proposed for the Flexible Job Shop Scheduling Problem with Automated Guided Vehicle (FJSP-AGV). The method converts FJSP-AGV into a Markov Decision Process (MDP) and generates an optimal scheduling policy using an improved DDQN.
JOURNAL OF INTELLIGENT MANUFACTURING
(2023)
Article
Automation & Control Systems
Cong Luo, Wenyin Gong, Rui Li, Chao Lu
Summary: In this paper, a knowledge-driven multi-objective evolutionary algorithm is proposed to solve the distributed permutation flow shop problem in heterogeneous factories. The algorithm aims to minimize the makespan and total energy consumption, and incorporates several improvements to enhance its effectiveness.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Review
Computer Science, Information Systems
Mei Li, Gai-Ge Wang
Summary: The manufacturing industry plays a crucial role in a country's productivity level and economic development, but it also brings about environmental pollution and resource scarcity. Therefore, researching Green Shop Scheduling Problems (GSSPs) has become an important topic, aiming to reduce resource consumption and environmental pollution while achieving economic benefits through behavior control. In the context of Industry 4.0, GSSPs require re-examination and study.
INFORMATION SCIENCES
(2022)
Article
Computer Science, Artificial Intelligence
Hao Wang, Junfu Cheng, Chang Liu, Yuanyuan Zhang, Shunfang Hu, Liangyin Chen
Summary: This research proposes a new dynamic multi-objective flexible job shop scheduling problem and designs a scheduling algorithm based on deep reinforcement learning. Experimental results demonstrate that the algorithm outperforms other methods in terms of performance improvement.
APPLIED SOFT COMPUTING
(2022)
Article
Computer Science, Hardware & Architecture
Yan Lei, Huan Xie, Tao Zhang, Meng Yan, Zhou Xu, Chengnian Sun
Summary: This article proposes a feature-based fault localization methodology (Feature-FL) that evaluates the suspiciousness of faulty statements by combining the diversity of program features. Experimental results show that Feature-FL outperforms spectrum-based formulas in real fault scenarios.
IEEE TRANSACTIONS ON RELIABILITY
(2022)
Article
Computer Science, Information Systems
Jian Hu, Huan Xie, Yan Lei, Ke Yu
Summary: This paper proposes a light-weight data augmentation method called Lamont to improve the effectiveness of original FL methods and the efficiency of Aeneas. Lamont uses revised LDA to reduce the dimensionality of the coverage matrix and employs SMOTE to generate synthesized failing tests. Experimental results show that Lamont outperforms original FL methods and is more efficient than Aeneas with comparable effectiveness.
INFORMATION AND SOFTWARE TECHNOLOGY
(2023)
Proceedings Paper
Computer Science, Software Engineering
Yan Lei, Chunyan Liu, Huan Xie, Sheng Huang, Meng Yan, Zhou Xu
Summary: This study proposes a data augmentation approach called BCL-FL based on between-class learning to address the problem of imbalanced data in fault localization. The experimental results show that BCL-FL significantly improves the effectiveness of existing FL techniques.
2022 IEEE INTERNATIONAL CONFERENCE ON SOFTWARE ANALYSIS, EVOLUTION AND REENGINEERING (SANER 2022)
(2022)
Proceedings Paper
Computer Science, Software Engineering
Junji Yu, Yan Lei, Huan Xie, Lingfeng Fu, Chunyan Liu
Summary: Automated fault localization techniques use runtime information to identify suspicious statements that may be responsible for program failures. However, coincidental correctness (CC) affects the effectiveness of fault localization. To address this issue, researchers propose a context-based cluster fault localization approach (CBCFL) that incorporates failure context into cluster analysis to improve the identification of CC tests.
30TH IEEE/ACM INTERNATIONAL CONFERENCE ON PROGRAM COMPREHENSION (ICPC 2022)
(2022)
Proceedings Paper
Computer Science, Software Engineering
Huan Xie, Yan Lei, Meng Yan, Yue Yu, Xin Xia, Xiaoguang Mao
Summary: Aeneas is a universal data augmentation approach proposed for fault localization. It generates synthesized failing test cases from a reduced feature space using a revised principal component analysis and a conditional variational autoencoder, improving the accuracy and effectiveness of fault localization.
2022 ACM/IEEE 44TH INTERNATIONAL CONFERENCE ON SOFTWARE ENGINEERING (ICSE 2022)
(2022)
Proceedings Paper
Computer Science, Software Engineering
Deheng Yang, Yan Lei, Xiaoguang Mao, David Lo, Huan Xie, Meng Yan
Summary: This paper introduces DEPTEST, an automated DatasEt Purification technique that leverages coverage analysis and delta debugging to identify and filter out irrelevant code changes from bug datasets. The experiment on the widely used Defects4J dataset shows that even in well-isolated bug fixes, 41.01% of human-written patches can be further reduced, demonstrating the potential of DEPTEST in aiding the construction of accurate bug fix datasets.
2021 IEEE INTERNATIONAL CONFERENCE ON SOFTWARE ANALYSIS, EVOLUTION AND REENGINEERING (SANER 2021)
(2021)
Article
Computer Science, Interdisciplinary Applications
Rafael Praxedes, Teobaldo Bulhoes, Anand Subramanian, Eduardo Uchoa
Summary: The Vehicle Routing Problem with Simultaneous Pickup and Delivery is a classical optimization problem that aims to determine the least-cost routes while meeting pickup and delivery demands and vehicle capacity constraints. In this study, a unified algorithm is proposed to solve multiple variants of the problem, and extensive computational experiments are conducted to evaluate the algorithm's performance.
COMPUTERS & OPERATIONS RESEARCH
(2024)
Article
Computer Science, Interdisciplinary Applications
Ragheb Rahmaniani, Teodor Gabriel Crainic, Michel Gendreau, Walter Rei
Summary: Benders decomposition (BD) is a popular solution algorithm for stochastic integer programs. However, existing parallelization methods often suffer from inefficiencies. This paper proposes an asynchronous parallel BD method and demonstrates its effectiveness through numerical studies and performance enhancement strategies.
COMPUTERS & OPERATIONS RESEARCH
(2024)
Article
Computer Science, Interdisciplinary Applications
Giulia Caselli, Maxence Delorme, Manuel Iori, Carlo Alberto Magni
Summary: This study addresses a real-world scheduling problem and proposes four exact methods to solve it. The methods are evaluated through computational experiments on different types of instances and show competitive advantages on specific subsets. The study also demonstrates the generalizability of the algorithms to related scheduling problems with contiguity constraints.
COMPUTERS & OPERATIONS RESEARCH
(2024)
Article
Computer Science, Interdisciplinary Applications
Shaowen Yao, Chao Tang, Hao Zhang, Songhuan Wu, Lijun Wei, Qiang Liu
Summary: This paper examines the problem of two-dimensional irregular multiple-size bin packing and proposes a solution that utilizes an iteratively doubling binary search algorithm to find the optimal bin combination, and further optimizes the result through an overlap minimization approach.
COMPUTERS & OPERATIONS RESEARCH
(2024)
Article
Computer Science, Interdisciplinary Applications
Decheng Wang, Ruiyou Zhang, Bin Qiu, Wenpeng Chen, Xiaolan Xie
Summary: Consideration of driver-related constraints, such as mandatory work break, in vehicle scheduling and routing is crucial for safety driving and protecting the interests of drivers. This paper addresses the drop-and-pull container drayage problem with flexible assignment of work break, proposing a mixed-integer programming model and an algorithm for solving realistic-sized instances. Experimental results show the effectiveness of the proposed algorithm in handling vehicle scheduling and routing with work break assignment.
COMPUTERS & OPERATIONS RESEARCH
(2024)
Article
Computer Science, Interdisciplinary Applications
William N. Caballero, Jose Manuel Camacho, Tahir Ekin, Roi Naveiro
Summary: This research provides a novel probabilistic perspective on the manipulation of hidden Markov model inferences through corrupted data, highlighting the weaknesses of such models under adversarial activity and emphasizing the need for robustification techniques to ensure their security.
COMPUTERS & OPERATIONS RESEARCH
(2024)
Article
Computer Science, Interdisciplinary Applications
Davood Zaman Farsa, Shahryar Rahnamayan, Azam Asilian Bidgoli, H. R. Tizhoosh
Summary: This paper proposes a multi-objective evolutionary framework for compressing feature vectors using deep autoencoders. The framework achieves high classification accuracy and efficient image representation through a bi-level optimization scheme. Experimental results demonstrate the effectiveness and efficiency of the proposed framework in image processing tasks.
COMPUTERS & OPERATIONS RESEARCH
(2024)
Article
Computer Science, Interdisciplinary Applications
Matthew E. Scherer, Raymond R. Hill, Brian J. Lunday, Bruce A. Cox, Edward D. White
Summary: This paper discusses instance generation methods for the multidemand multidimensional knapsack problem and introduces a primal problem instance generator (PPIG) to address feasibility issues in current instance generation methods.
COMPUTERS & OPERATIONS RESEARCH
(2024)
Article
Computer Science, Interdisciplinary Applications
Yin Yuan, Shukai Li, Lixing Yang, Ziyou Gao
Summary: This paper investigates the design of real-time train regulation strategies for urban rail networks to reduce train deviations and passenger waiting times. A mixed-integer nonlinear programming (MINLP) model is used and an efficient iterative optimization (IO) approach is proposed to address the complexity. The generalized Benders decomposition (GBD) technique is also incorporated. Numerical experiments show the effectiveness and computational efficiency of the proposed method.
COMPUTERS & OPERATIONS RESEARCH
(2024)
Article
Computer Science, Interdisciplinary Applications
Xinghai Guo, Netirith Narthsirinth, Weidan Zhang, Yuzhen Hu
Summary: This study proposes a bi-level scheduling method that utilizes unmanned surface vehicles for container transportation. By formulating mission decision and path control models, efficient container transshipment and path planning are achieved. Experimental results demonstrate the effectiveness of the proposed approach in guiding unmanned surface vehicles to complete container transshipment tasks.
COMPUTERS & OPERATIONS RESEARCH
(2024)
Review
Computer Science, Interdisciplinary Applications
Jose-Fernando Camacho-Vallejo, Carlos Corpus, Juan G. Villegas
Summary: This study aims to review the published papers on implementing metaheuristics for solving bilevel problems and performs a bibliometric analysis to track the evolution of this topic. The study provides a detailed description of the components of the proposed metaheuristics and analyzes the common combinations of these components. Additionally, the study provides a detailed classification of how crucial bilevel aspects of the problem are handled in the metaheuristics, along with a discussion of interesting findings.
COMPUTERS & OPERATIONS RESEARCH
(2024)
Article
Computer Science, Interdisciplinary Applications
Xudong Diao, Meng Qiu, Gangyan Xu
Summary: In this study, an optimization model for the design of an electric vehicle-based express service network is proposed, considering limited recharging resources and power management. The proposed method is validated through computational experiments on realistic instances.
COMPUTERS & OPERATIONS RESEARCH
(2024)
Article
Computer Science, Interdisciplinary Applications
Ramon Piedra-de-la-Cuadra, Francisco A. Ortega
Summary: This study proposes a procedure to select candidate sites optimally for ensuring energy autonomy and reinforced service coverage for electric vehicles, while considering demand and budget restrictions.
COMPUTERS & OPERATIONS RESEARCH
(2024)
Article
Computer Science, Interdisciplinary Applications
Danny Blom, Christopher Hojny, Bart Smeulders
Summary: This paper focuses on a robust variant of the kidney exchange program problem with recourse, and proposes a cutting plane method for solving the attacker-defender subproblem. The results show a significant improvement in running time compared to the state-of-the-art, and the method can solve previously unsolved instances. Additionally, a new practical policy for recourse is proposed and its tractability for small to mid-size kidney exchange programs is demonstrated.
COMPUTERS & OPERATIONS RESEARCH
(2024)
Article
Computer Science, Interdisciplinary Applications
Anqi Li, Congying Han, Tiande Guo, Bonan Li
Summary: This study proposes a general framework for designing linear programming instances based on the preset optimal solution, and validates the effectiveness of the framework through experiments.
COMPUTERS & OPERATIONS RESEARCH
(2024)