Article
Engineering, Industrial
Pengfei Wang, Zhenkun Xu, Diyi Chen
Summary: Condition-based maintenance (CBM) is crucial for intelligent operation and maintenance of hydropower generation units (HPGUs) at hydropower stations. However, there is a lack of comprehensive research on an integrated theoretical framework for CBM of HPGUs based on reliability assessment. In this study, we propose a theoretical framework that combines the generalized proportional hazard model and semi-Markov decision process to achieve a balance between reliability prediction and optimization. The effectiveness of the method is demonstrated through actual cases, which show significant cost savings and improved maintenance cycle accuracy compared to traditional models. The model's performance under variable operating conditions is also discussed, further enhancing the framework's completeness.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2023)
Article
Engineering, Industrial
Rebaiaia Mohamed-Larbi, Ait-Kadi Daoud
Summary: This paper proposes a Novel Condition-based maintenance (CBM) decision aid model for optimizing the maintenance of complex multi-component systems. The model aims to predictively remove components before failure by determining measurements on the components, in order to optimize maintenance costs and spare parts usage.
INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH
(2023)
Article
Computer Science, Artificial Intelligence
Ram S. Mohril, Bhupendra S. Solanki, Makarand S. Kulkarni, Bhupesh K. Lad
Summary: Accurate maintenance decision making is crucial for organizations, but the impact of human error on equipment life is still unknown. This paper presents a machine learning approach to estimate the residual life of a component while considering human error in maintenance. The developed model effectively handles uncertainties and reduces mathematical complexity.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Engineering, Mechanical
Mengyao Gu, Jiangqin Ge
Summary: The study introduces a novel SbRLP method based on compound similarity, which measures global and local similarities using weighted Hausdorff distance and cosine adjustment similarity, combines them using the control coefficient U, and achieves more accurate RUL prediction.
JOURNAL OF MECHANICAL SCIENCE AND TECHNOLOGY
(2022)
Article
Engineering, Marine
Yuandong Wang, Baoping Cai, Yanping Zhang, Jing Liu, Javed Akbar Khan, Yiliu Liu, Rongkang Li, Zhengde Chu, Zengkai Liu, Yonghong Liu
Summary: This paper proposes a condition-based maintenance (CBM) method based on remaining useful life (RUL) prediction, which dynamically sets inspection time and optimizes maintenance cost by considering maintenance preparation time and RUL prediction results.
Article
Engineering, Industrial
Jun Xu, Zhenglin Liang, Yan-Fu Li, Kaibo Wang
Summary: This paper investigates the optimal CBM policy for a K-out-of-N: G system under periodic inspection, with focus on economic dependency, stochastic dependency, and imperfect maintenance to minimize expected long-run discounted cost. Utilizing Markov decision process framework, the study discretizes state space and employs value iteration algorithm with Monte Carlo simulation for optimization. Numerical studies on a 1-out-of-2: G system examine impacts of degradation processes, copula functions, and imperfect maintenance, providing insights for multi-component system maintenance and conducting sensitivity analysis on cost-related parameters.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2021)
Article
Physics, Multidisciplinary
Sudheesh Kumar Kattumannil, E. P. Sreedevi, Narayanaswamy Balakrishnan
Summary: In this paper, a generalized measure of cumulative residual entropy is introduced and its properties are studied. Relationships between different measures of entropy and extropy are established.
Article
Engineering, Industrial
Yan-Hui Lin, Ze-Qi Ding, Yan-Fu Li
Summary: Similarity based methods directly measure the similarities between degradation trajectories to predict remaining useful life (RUL), suitable for scenarios without known equipment operation beginning time or established degradation models. However, determining appropriate similarity measures can be laborious, and these methods lack evaluation of prediction uncertainty. This paper proposes a Gaussian process (GP) modeling framework that enriches similarity measures and improves interpretability, providing both predicted RUL values and associated prediction uncertainty. A novel active learning method is also proposed to alleviate computational complexity, enabling real-time application. The effectiveness of the proposed framework is demonstrated through case studies.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2023)
Article
Engineering, Industrial
Michiel A. J. Uit Het Broek, Ruud H. Teunter, Bram de Jonge, Jasper Veldman
Summary: Condition-based maintenance and production policies play important roles in manufacturing industry as they can reduce costs and improve reliability by dynamically adjusting production rates and implementing maintenance. Integrating maintenance and production policies together can effectively lower failure risks and achieve higher cost savings.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2021)
Article
Engineering, Mechanical
He-Fei Li, Jia-Liang Wei, Shao-Hua Li, Yong-Qiang Liu, Xiao-Hui Gu, Ze-Chao Liu, Shao-Pu Yang
Summary: The study investigated the tensile and fatigue properties of high-speed train bearings after different operating mileages. Experimental results showed that the microhardness, tensile strength, elongation, and fatigue strength of the bearing steel decreased with increasing operating mileage. Based on Miner's law, a generalized linear cumulative damage theory was proposed to linearly superimpose fatigue damage per unit mileage. Consequently, a residual strength model was established to evaluate the fatigue life of the bearings, providing new insights for fatigue life prediction under random loadings.
FATIGUE & FRACTURE OF ENGINEERING MATERIALS & STRUCTURES
(2023)
Article
Operations Research & Management Science
Stefanos Leonardos, Constandina Koki, Costis Melolidakis
Summary: In this paper, the unimodality conditions for distributions that describe markets with stochastic demand are studied. The paper analyzes game-theoretic models of market competition and supply chain coordination and characterizes optimal prices as fixed points of the mean residual life function of the demand distribution. It finds that markets with increasing price elasticity of demand correspond to distributions with decreasing generalized mean residual life and finite second moment.
ANNALS OF OPERATIONS RESEARCH
(2023)
Article
Computer Science, Interdisciplinary Applications
Longjie Li, Hui Wang, Shiyu Fang, Na Shan, Xiaoyun Chen
Summary: This paper proposes a supervised similarity-based link prediction method, which measures the connection likelihood of a node pair by extracting structural features and searching for positive and negative k-nearest neighbors. Experimental results demonstrate that the proposed method outperforms other methods in terms of accuracy and stability.
INTERNATIONAL JOURNAL OF MODERN PHYSICS C
(2021)
Article
Engineering, Mechanical
Kai-Shang Li, Run-Zi Wang, Lv-Yi Cheng, Ti-Wen Lu, Xian-Cheng Zhang, Shan-Tung Tu, Guo-Dong Zhang, Zhi-Chao Fan
Summary: In this study, a dislocation-based crystal plasticity finite element (CPFE) framework was used to investigate the effects of loading conditions on dwell-fatigue crack initiation life. A life prediction approach based on accumulated energy dissipation at half-life cycle was proposed, and a new three-dimensional damage tolerance diagram was introduced to evaluate the residual dwell fatigue life.
INTERNATIONAL JOURNAL OF FATIGUE
(2022)
Article
Engineering, Mechanical
Chaotao Liu, Ye Song, Fansong Li, Pingbo Wu, Yunguang Ye
Summary: Metro bogie frames occasionally experience fatigue failure during operation. Through fatigue fractography, it is determined that the fusion defect of the weld root at the frame's hot spot is the internal cause of fatigue failure. Based on field testing, it is found that the modal frequency of the frame coincides with the wheel-rail coupling vibration frequency under specific track structures, which is the external cause of the frame's fatigue failure. The wheel out-of-roundness (OOR) and the curved track with a radius below 600 m also significantly impact the fatigue life. The evaluation of the residual life provides a reference for optimizing the maintenance strategy of the bogie frame.
ENGINEERING FAILURE ANALYSIS
(2023)
Article
Engineering, Manufacturing
Shu-Lian Xie, Feng Xue, Wei-Min Zhang, Jia-Wei Zhu
Summary: This study proposes a maintenance policy considering reconfiguration to improve system reliability and availability, reduce downtime and maintenance costs by optimizing the concurrent maintenance time window of devices and determining whether to reconfigure them.
ADVANCES IN MANUFACTURING
(2023)