Defining and implementing a distributed and reconfigurable information system for prognostics

A Novel Predictive Selective Maintenance Strategy Using Deep Learning and Mathematical Programming

Organizations strive to maximize efficiency in their manufacturing processes, yet they must also consider broader repercussions, as industrial activity directly impacts the environment and society. The adoption of innovative technologies and initiatives to mitigate this impact is therefore essential. Traditional asset maintenance plays a critical role in ensuring high equipment availability, but there is a clear need to evolve toward predictive and sustainable maintenance strategies to enhance reliability, safety, and equipment lifespan. This shift redefines maintenance itself, aligning it with circular economy principles and Industry 4.0 solutions to prevent unplanned downtime, reduce failures, and improve personnel and facility safety. This research examines the transition from traditional preventive maintenance to predictive and sustainable maintenance in a real-world industrial context, comparing the design of a neural network with the ARIMAX technique to develop reliable predictive models. The study aims to facilitate a paradigm shift by proposing a predictive model that reduces unplanned shutdowns and optimizes spare parts and labor utilization. The practical application focuses on a hydrogen compressor in the petrochemical industry, demonstrating the model’s potential for operational and sustainability improvements.

Modern maintenance strategies, such as predictive and prescriptive maintenance, which derived from the concept of Industry and Maintenance 4.0, involve the application of the Industrial Internet of Things (IIoT) to connect maintenance objects enabling data collection and analysis that can help make better decisions on maintenance activities. Data collection is the initial step and the foundation of any modern Predictive or Prescriptive maintenance strategy because it collects data that can then be analysed to provide useful information about the state of maintenance objects. Condition monitoring of rotary equipment is one of the most popular maintenance methods because it can distinguish machine state between multiple fault types. The topic of this paper is the presentation of an automated system for data collection, processing and interpretation of rotary equipment state that is based on IIoT framework consisting of an IIoT accelerometer, edge and fog devices, web API and database. Additionally, ISO 10816-1 guidance has been followed to develop module for evaluation of vibration severity. The collected data is also visualized in a dashboard in a near-real time and shown to maintenance engineering, which is crucial for pattern monitoring. The developed system was launched in laboratory conditions using rotating equipment failure simulator to test the logic of data collection and processing. A proposed system has shown that it is capable of automated periodic data collection and processing from remote places which is achieved using Node RED programming environment and MQTT communication protocol that enables reliable, lightweight, and secure data transmission.

Effective predictive maintenance (PdM) strategy is needed in microelectronic manufacturing to reduce cost and loss of cycle time associated with unplanned maintenance events in fab. Adoption of recent advances in big data analytics methods will help in developing an effective PdM strategy. This paper discusses advanced machine learning based PdM strategy. Various approaches to formulate the PdM problem and solve it are discussed. The performance of respective machine learning algorithms on equipment data is evaluated. To benefit from the strengths of various algorithms a super learning based approach with deep neural networks is proposed. This approach is robust compared to single method selection and handles changing production environment well. This cross validation based stacked model ensemble predictive maintenance strategy is also easy to automate.

Maintenance, production and quality are strongly linked to each other. Predictive maintenance is an effective way to eliminate the potential failures and ensure the stable operation of the manufacturing system, and further improve the reliability of the manufacturing system and the quality of manufactured products. Regarding the intelligent manufacturing philosophy “prediction and manufacturing#x201D;, a novel predictive maintenance strategy is proposed, in which the product quality control is integrated into predictive maintenance decision-making. First, the key process variables that characterize equipment wear are identified and integrated into the modeling of the equipment failure rate. Second, quality deviation that characterizes product quality level is defined based on co-effect between manufacturing system component reliability and product quality (i.e., Q-R chain). Third, the optimal maintenance strategy is obtained by optimizing the quality cost, maintenance cost, and interruption cost simultaneously. Finally, a case study on the predictive maintenance strategy decision-making for a cylinder head manufacturing system is presented to illustrate the effectiveness of the proposed method. The final result shows that the proposed method can achieve a superior economic performance relative to the conventional preventive maintenance mode.

The increasing complexity of electrical power systems necessitates advanced fault detection and predictive maintenance strategies to enhance operational efficiency and grid reliability. Traditional maintenance approaches, such as reactive and preventive maintenance, have proven insufficient in mitigating unplanned outages and optimizing asset utilization. Recent advancements in artificial intelligence (AI) have introduced data-driven solutions that significantly improve fault classification, failure prediction, and automated recovery processes. This study conducts a systematic review of 180 high-quality peer-reviewed articles, following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to ensure a rigorous and transparent research methodology. The findings reveal that AI-driven predictive maintenance methods, including machine learning, deep learning, digital twin technology, IoT-enabled sensor networks, and self-healing grids, have outperformed traditional fault detection techniques in terms of accuracy, adaptability, and cost-effectiveness. AI-based fault detection models achieve an average accuracy of 85% to 95%, reducing false alarms by 50% and minimizing power restoration times by up to 60%. The integration of IoT sensors with real-time analytics has improved anomaly detection rates by 28%, while digital twin technology has enhanced predictive maintenance efficiency, reducing unplanned outages by 35%. Additionally, self-healing grid mechanisms, powered by reinforcement learning algorithms, have demonstrated the ability to autonomously isolate faults and reconfigure energy distribution, preventing nearly 45% of potential service disruptions. Despite these advancements, challenges such as the black-box nature of deep learning models, cybersecurity vulnerabilities, and interoperability with legacy systems continue to pose barriers to large-scale AI adoption. The study highlights the need for explainable AI frameworks, standardized data governance policies, and enhanced cybersecurity measures to ensure the sustainable deployment of AI in power grid management. The findings provide valuable insights for researchers, utility companies, and policymakers seeking to enhance the resilience and efficiency of modern electrical power systems through AI-driven fault detection and predictive maintenance strategies.

With the evolution of fifth generation (5G) network technologies, network maintenance strategies have become increasingly complex, necessitating the use of predictive analysis enabled by Machine Learning (ML) algorithms. This paper emphasizes exploring how ML algorithms can further enhance predictive maintenance in 5G and future networks. It reviews the current literature on this interdisciplinary topic, identifying key ML models such as Decision Trees, Neural Networks, and Support Vector Machines, and discussing their benefits and limitations. Special attention is given to the methodologies in applying these models, handling of data stages, and the training process. Major challenges in implementing ML in the context of network maintenance, such as data privacy, data gathering, model training, and generalizability, are discussed. Furthermore, the research aims to go beyond predicting maintenance needs to introduce a proactive approach in improving overall network performance and pre-empting potential issues based on ML predictions. The paper also discusses possible future trends including advancements in ML algorithms, Automated Machine Learning (AutoML), Explainable AI, and others. The objective is to provide a comprehensive understanding of the current ML-based predictive maintenance field and outline possibilities for future research. The study finds that the application of ML algorithms continues to show promise in transforming the landscape of network management by improving predictive maintenance and proactive performance enhancement strategies. It remains a challenging yet important area in the context of 5G networks.

The digitalization of industrial maintenance has transformed traditional approaches to asset management, particularly in the electrical domain. The integration of smart indicators, predictive analytics, and decision support systems has enabled industries to move from reactive and preventive models toward predictive and prescriptive maintenance strategies. This paradigm shift not only enhances operational efficiency but also extends equipment lifetime, reduces downtime, and strengthens safety and energy management. The purpose of this paper is to analyze how digital tools and intelligent indicators reshape the management of industrial utilities, emphasizing the interplay between predictive maintenance, data analytics, and decision-making. By reviewing recent academic literature, the article demonstrates that the successful adoption of smart indicators requires both robust technical architectures and human-centered decision support frameworks, highlighting challenges and opportunities for future advancements in the field.

A survey of deep learning-driven architecture for predictive maintenance

Abstract—Predictive maintenance (PdM) has emerged as a transformative approach in semiconductor manufacturing, leveraging IoT sensor data to enhance equipment reliability and optimize production processes. This review explores the implementation of predictive maintenance strategies in semiconductor manufacturing, focusing on the role of IoT sensor networks, advanced analytics techniques, and integration with Manufacturing Execution Systems (MES). IoT sensor networks enable comprehensive real-time data collection on equipment performance and environmental conditions, thereby providing a foundation for predictive maintenance. Time series forecasting algorithms, such as ARIMA, exponential smoothing, and machine learning-based approaches, are employed to anticipate potential equipment failures. Anomaly detection techniques, including statistical methods and machine-learning algorithms, are used to identify unusual patterns or behaviors that are indicative of impending issues. The integration of predictive maintenance insights with MES allows for real-time decision making, process optimization, and improved overall equipment effectiveness. However, challenges persist in terms of data quality, scalability, and cybersecurity, requiring ongoing research and industry collaboration. Early adopters reported significant reductions in unplanned downtime, optimized maintenance schedules, and improved product quality. As the semiconductor industry continues to evolve, predictive maintenance is expected to play a crucial role in maintaining competitiveness and meeting the growing demand. Further research, standardization efforts, and development of best practices are essential to fully realize the potential of predictive maintenance in semiconductor manufacturing. Keywords—predictive maintenance (PdM), semiconductor manufacturing, IoT sensor data, equipment reliability, forecasting, anomaly detection, unplanned downtime, product quality

Sutirtha Chatterjee) join the editor-in-chief (Robert Davison) to craft a position statement regarding the ISJ's view on theory. It is applicable, with sensitivity, to the empirical research articles that we consider for publication.

Maximizing the value of OEE is a crucial thing to do for a company in order to increase their productivity. OEE is one of the most common technique to measure equipment’s effectiveness. To enhance the number of OEE, it is very important to do a maintenance task to keep the equipment in a good condition. Predictive maintenance task is one of the maintenance strategies to see any potential failure in an equipment and do a preventive way to prevent it from a total stop/break down. As a result, understanding the needs to do a maintenance upgrades for better performance and proposing a good predictive maintenance strategy will enhance the value of OEE and result in cost savings. The objective of this paper is to propose the necessary predictive maintenance task using analytical hierarchy process (AHP) using three respondents to decide the AHP process and consistency ratio to get the most suitable maintenance task for fertilizer company as small medium enterprise.

Offshore platforms are vital assets for the oil and gas industry, serving as the primary facilities for exploration, extraction, and processing. Maintenance logistics plays a crucial role in ensuring these platforms operate efficiently and safely. However, the remote and harsh environments of offshore platforms present significant challenges for maintenance activities. Traditional maintenance strategies often struggle to meet the demands of these environments, leading to inefficiencies, increased costs, and potential safety risks. This review discusses the application of Artificial Intelligence (AI) in optimizing maintenance logistics on offshore platforms. Current strategies involve a combination of preventive, predictive, and corrective maintenance approaches. Preventive maintenance schedules regular inspections and replacements based on predetermined intervals, while predictive maintenance utilizes data analytics to predict equipment failures and plan maintenance activities accordingly. Corrective maintenance addresses issues as they arise, often in response to unexpected failures. AI offers opportunities to enhance these strategies by leveraging advanced data analytics, machine learning, and optimization algorithms. AI-enabled predictive maintenance can analyze vast amounts of data from sensors, historical maintenance records, and environmental factors to forecast equipment failures with greater accuracy. This allows for proactive maintenance planning, minimizing downtime and reducing maintenance costs. Furthermore, AI can optimize maintenance logistics by improving resource allocation and scheduling. Through real-time monitoring and analysis, AI systems can prioritize maintenance tasks based on urgency, equipment criticality, and resource availability. This ensures that maintenance crews are deployed efficiently, reducing idle time and improving overall productivity. Future innovations in AI for maintenance logistics on offshore platforms include the integration of Internet of Things (IoT) devices and autonomous systems. IoT sensors can provide real-time data on equipment condition and environmental factors, enabling more precise predictive maintenance models. Autonomous maintenance robots equipped with AI algorithms can perform routine inspections and minor repairs, reducing the need for human intervention in hazardous environments. However, implementing AI in offshore maintenance logistics also poses challenges, including data quality, cybersecurity, and workforce readiness. Ensuring data accuracy and reliability is crucial for effective AI models, requiring robust data collection and management processes. Cybersecurity measures must be strengthened to protect AI systems from malicious attacks that could disrupt operations or compromise safety. Additionally, workforce training and education are essential to prepare personnel for working alongside AI systems and interpreting AI-generated insights. Optimizing maintenance logistics on offshore platforms with AI offers significant benefits in terms of efficiency, cost savings, and safety. By leveraging AI technologies, current maintenance strategies can be enhanced, and future innovations can revolutionize offshore maintenance practices, making operations more sustainable and resilient in the face of evolving challenges.

Monitoring and control of air filtration systems: Digital twin based on 1D computational fluid dynamics simulation and experimental data

The challenges facing providers of global communication and information systems are discussed. Globe-girdling systems are compared to more confined systems. Unique facets of global systems are illustrated using IBM's worldwide information system as an example of an integrated enterprise network. Progress is sketched in other uses, including travel reservation, banking and finance, military, research and education. It is concluded that global systems' evolution can be predicted, in large measure, from prior progress in shorter-reach applications.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>