Publication Search

The implementation of predictive maintenance supported by SAP Plant Maintenance (SAP PM) at PT Xyz has proven to be effective in reducing machine downtime, lowering maintenance costs, and improving asset reliability. The integration of SAP PM with Industry 4.0 technologies such as IoT sensors, AI-based analytics, and real-time notification systems strengthens operational efficiency and ensures continuous performance. Empirical results show improvements in key performance indicators, including a 20-25% reduction in downtime, a 30% reduction in maintenance costs, an increase in asset availability to 97%, an MTBF extension of up to 511 hours, and an OEE rate of 92.1%. These findings highlight the strategic role of digital predictive maintenance in increasing competitiveness and supporting long-term sustainability in manufacturing operations.

Predictive maintenance (PdM) plays a crucial role in modern industrial systems by minimizing downtime, reducing maintenance costs, and optimizing asset performance. However, many predictive models operate as “black box” systems, limiting transparency and making it difficult for operators to interpret their outputs. This study aims to integrate Explainable Artificial Intelligence (XAI) techniques with Remaining Useful Life (RUL) prediction models to improve both accuracy and interpretability. Various machine learning and deep learning approaches, including Support Vector Machines (SVM), Random Forest (RF), XGBoost, Long Short-Term Memory (LSTM), and Convolutional Neural Networks (CNN), are employed to predict RUL using real-time sensor data from rotating machinery. XAI methods such as SHAP, LIME, and attention mechanisms are applied to provide human-understandable explanations of model predictions. The models are evaluated based on accuracy, Root Mean Square Error (RMSE), and interpretability scores. The results show that XAI-enhanced models outperform traditional approaches in predictive performance while offering greater transparency. These explanations help maintenance engineers better understand the factors influencing predictions, thereby improving decision-making and trust in the system. Nevertheless, the integration of XAI introduces additional computational complexity, which may pose challenges for large-scale industrial implementation. Overall, this study highlights the potential of combining XAI with RUL prediction to develop more reliable, transparent, and effective predictive maintenance solutions.

The Low Voltage Main Distribution Panel (LVMDP) is a critical component in industrial power distribution systems, functioning to regulate, control, and distribute electrical energy to various production equipment. During operation, LVMDP panels often operate under high electrical loads, which may lead to temperature increases in their components. Undetected temperature rise can result in performance degradation, equipment failure, and even fire hazards. Therefore, an effective monitoring method is required to detect the condition of electrical components at an early stage. This study aims to analyze the temperature difference (ΔT) of LVMDP components using the Infrared Thermography method as part of predictive maintenance. The research employs a quantitative descriptive approach with data collected through direct observation from July 1 to July 31 at PT. Dongjin Indonesia. The data consist of hotspot and ambient temperatures measured from several panel components, which are then analyzed to calculate the temperature difference (ΔT) as an indicator of component operating conditions. The results indicate that the highest temperature difference is 26.5 °C in the capacitor bank, while the lowest is 4 °C in other components. All ΔT values are below the threshold limit of 50 °C, indicating that the LVMDP components are in safe operating conditions and do not require corrective actions. Thus, Infrared Thermography is proven to be an effective method for early detection of component conditions and can enhance the reliability and safety of industrial power distribution systems.

This research investigates intelligent cooling system design for main ship engines operating in tropical waters, integrating advanced machinery engineering with human factors to address thermal management challenges affecting engine performance, reliability, and crew operational effectiveness. Tropical maritime environments impose severe cooling demands through elevated seawater temperatures (28-32°C), high ambient conditions (28-35°C), and accelerated biofouling, reducing conventional cooling system effectiveness by 15-25% while increasing maintenance burdens and operational risks. Through qualitative analysis involving marine engineers, chief engineers with tropical operational experience, cooling system manufacturers, naval architects, automation specialists, and maritime training institutions, this study examines how intelligent cooling systems incorporating variable-speed pumps, adaptive control algorithms, predictive maintenance, and crew-centered interfaces can optimize thermal management while supporting effective human-machine collaboration. Results demonstrate that intelligent systems can reduce cooling energy consumption by 20-35%, improve temperature stability by 50-65%, extend maintenance intervals by 40-80%, and enhance crew situational awareness through intuitive monitoring interfaces, while requiring comprehensive training programs developing technical understanding and operational competencies. Key implementation challenges include control system complexity, sensor reliability in harsh marine environments, integration with existing engine management platforms, crew competency development requirements, and lifecycle cost justification. Findings reveal that successful intelligent cooling system implementation requires holistic sociotechnical approach addressing machinery engineering optimization, automation technology deployment, and human capability development through coordinated design and training strategies. This research contributes to marine engineering literature by providing integrated frameworks for intelligent system design incorporating machinery performance, automation capabilities, and human factors supporting operational excellence in tropical maritime operations.

This study aims to analyze the transformative impact of digitalization and technology in the manufacturing sector on improving operational efficiency, particularly in budgeting and resource utilization, as well as to identify the main barriers to technology adoption. Using a Literature Review and Case Study Analysis of secondary data (journals, company reports, and industry publications), it was found that digitalization and Automation supported by Artificial Intelligence (AI) fundamentally transform budgeting functions. This transformation has been shown to improve budget accuracy by up to 50% (reducing human errors) and process efficiency by up to 25%, turning budgets from static documents into adaptive and predictive control tools. Positive impacts are also observed in operations through increased production capacity (revenue surge) and the implementation of Predictive Maintenance, which reduces expenditure and asset downtime, in line with the principles Cost Efficiency and Lean Manufacturing. Nevertheless, the adoption of advanced technology faces significant obstacles, namely high initial capital investment and skill gaps among the workforce. It is concluded that the success of digitalization heavily depends on strategic budget planning to overcome capital barriers and adequate allocation of funds for Human Resource (HR) training to support effective collaboration between humans and machines.

This study investigates the adoption of adaptive DevOps practices in embedded systems used in safety-critical industrial applications. Traditional DevOps models, which are primarily designed for cloud-based systems, face significant challenges when applied to embedded platforms due to hardware constraints, real-time performance requirements, and stringent safety standards. The research focuses on developing a tailored DevOps framework that integrates continuous integration/continuous delivery (CI or CD) pipelines, automation, real-time monitoring, and safety assurance processes to enhance system reliability, performance, and compliance with regulatory standards. The study uses a case study methodology, involving embedded system teams across multiple industrial sectors, to assess the impact of these adapted DevOps practices on system stability and operational efficiency. Key findings show that the adoption of adaptive DevOps practices led to significant improvements in system reliability, performance, and deployment stability. Continuous feedback mechanisms allowed for early issue detection and faster resolution, leading to enhanced system uptime and responsiveness. Additionally, the integration of safety assurance into the DevOps pipeline ensured that safety-critical systems complied with required safety integrity levels and certification standards. The study further explores the integration of DevOps with embedded safety-critical systems, highlighting the benefits of cross-domain collaboration, enhanced communication, and the ability to address the unique challenges of these platforms. The research also underscores the limitations of conventional DevOps models in embedded systems and presents practical implications for the wider adoption of DevOps in safety-critical industrial applications. Future research is recommended to refine DevOps frameworks for embedded systems, integrating emerging technologies like the Industrial Internet of Things (IIoT) and Digital Twins to further optimize performance, security, and predictive maintenance.

Background: The increasing complexity of industrial production systems requires machine condition monitoring solutions that are capable of operating in real time with high accuracy and responsiveness to support predictive maintenance strategies. Conventional cloud based monitoring systems often experience limitations such as high latency and dependence on stable network connectivity, which can delay decision making processes in critical industrial operations. Objective: This study aims to design and evaluate an Industrial Internet of Things (IIoT) architecture based on edge computing to improve the efficiency of industrial sensor data processing and accelerate anomaly detection in industrial machines. Method: The research adopts an experimental approach by designing a system architecture consisting of a sensor layer, edge computing layer, and cloud layer. Industrial sensors, including vibration, temperature, and current sensors, continuously collect machine operational data, which are then processed locally at the edge node using a machine learning based anomaly detection algorithm. System testing is conducted in a simulated manufacturing environment to evaluate performance based on latency, reliability, and detection accuracy. Results: The results indicate that edge based data processing significantly reduces latency compared with cloud-based processing and enables faster responses to machine condition changes. Additionally, the implemented anomaly detection algorithm achieves high accuracy in identifying abnormal sensor data patterns.

Background: The development of industrial technology in the Industry 4.0 era has encouraged the implementation of intelligent monitoring systems to improve machine reliability and operational efficiency. However, machine fault diagnosis systems based on artificial intelligence often face limitations in terms of interpretability because the models used are complex and difficult to explain. Objective: This study aims to develop a deep learning-based industrial machine fault diagnosis system integrated with an Explainable Artificial Intelligence (XAI) approach to improve diagnostic accuracy while providing interpretable insights for users. Method: The research method involves collecting data from industrial machine sensors consisting of vibration signals, temperature measurements, and acoustic signals, followed by data preprocessing and feature extraction processes. The processed data are then used to train a deep learning-based diagnostic model, after which explainability methods such as SHAP or LIME are applied to analyze the contribution of each feature to the model’s prediction results. Model performance is evaluated using accuracy, precision, recall, and F1-score metrics. Results: The results indicate that the proposed deep learning model achieves better performance compared to conventional machine learning methods such as Support Vector Machine and Random Forest. Furthermore, the explainability analysis reveals that vibration amplitude, increases in machine component temperature, and anomalies in acoustic signals are the main factors influencing machine fault detection. Therefore, the proposed system not only improves the accuracy of machine fault diagnosis but also provides transparency in the decision-making process, thereby supporting the implementation of predictive maintenance in smart manufacturing environments.

Mining operations rely heavily on the performance and reliability of heavy equipment used in the production process. One of the most important hauling units in open-pit mining is the dump truck, which functions to transport overburden and coal from the mining front to disposal areas. Due to high operational intensity, dump trucks require effective maintenance management to ensure equipment reliability and reduce unexpected downtime. However, maintenance activities are often carried out based only on routine service schedules without analytical planning based on historical data. This study aims to analyze the implementation of forecasting methods in maintenance management to improve the effectiveness of dump truck maintenance planning in mining operations. The research was conducted during field work practice at PT Putra Perkasa Abadi Jobsite BIB, Tanah Bumbu, South Kalimantan. The data used were historical maintenance records of dump truck units obtained from the maintenance department. The research method used a quantitative approach with time series forecasting analysis to identify maintenance patterns and estimate future maintenance needs. The results show that forecasting-based maintenance planning can help companies predict maintenance requirements more accurately and prepare maintenance resources more efficiently. Furthermore, the implementation of forecasting methods can reduce unexpected equipment failures and support operational efficiency in mining activities.

This study analyzes the maintenance process of the Roll Bending machine used in the wind tower production line at PT Kenertec Power System, Cilegon, Indonesia. The Roll Bending machine plays a crucial role in shaping steel plates into cylindrical shell components, which are later assembled into wind tower sections. The objective of this research is to identify maintenance patterns, types of failures, and improvement strategies to ensure machine reliability and operational efficiency. The research employed observation, interviews with maintenance personnel, and documentation review to collect relevant data. The findings show that the machine experienced multiple failures, mostly related to hydraulic system leaks, PLC programming errors, and component wear such as cylinders, seals, and gear pumps. A significant increase in corrective maintenance activities occurred between August 2023 and April 2024, particularly in February 2024, indicating the need for a more consistent predictive maintenance strategy. The implications of this study highlight that optimized maintenance scheduling and monitoring are essential to reduce downtime, avoid production delays, and maintain product quality. This research is expected to support maintenance decision-making and contribute to the improvement of industrial machine reliability in wind tower manufacturing operations.

This study aims to analyze production performance and the factors influencing the productivity of mining operations at PT. XYZ during August 2025. The evaluation covers production achievement against the corporate work plan (RKAP) and the owner’s operational plan, equipment availability (Physical Availability), the productivity of loading and hauling units, and various types of loss time that reduce effective working hours. The results indicate that production realization reached only 65% of the RKAP target, while achieving 102% of the owner’s plan for total material. Low equipment availability, high loss time such as no hauler, wait operator, and front preparation, as well as the underperformance of 80-ton and 100-ton units, were identified as the main contributors to production deviation. Additional influencing factors include unit reassignment, suboptimal haul road conditions, and insufficient operational fleet numbers. The study recommends optimizing fleet management, enhancing preventive and predictive maintenance programs, reorganizing hauling workflows, and controlling dominant loss time sources to improve operational efficiency and production target achievement in future periods.

The reliability of power distribution systems is a crucial factor in ensuring stable electricity supply for industrial, commercial, and household users. Conventional protection systems often face limitations in terms of real-time monitoring, remote control, and adaptive responses to fault conditions, which can result in longer outage durations and higher operational costs. This research aims to develop a smart protection system for power distribution using Internet of Things (IoT) technology to enhance system reliability. The proposed method integrates IoT-enabled sensors, microcontrollers, and communication modules to monitor critical parameters such as voltage, current, and frequency in real time. Data are transmitted to a cloud-based platform for analysis and decision-making, enabling rapid detection of abnormalities and remote tripping of circuit breakers. The prototype was tested under various fault scenarios, including short circuits and overloads, and demonstrated faster response times compared to conventional systems. Results show that the IoT-based protection system improved fault detection accuracy, reduced downtime, and provided predictive maintenance insights through data analytics. The synthesis of these findings highlights that integrating IoT into protection mechanisms not only increases operational reliability but also supports the transition toward smart grids. In conclusion, the developed system proves effective in addressing the limitations of traditional protection systems by offering real-time monitoring, automation, and enhanced decision-making for modern power distribution networks.

The increasing integration and digitization of smart grid systems have exposed them to a variety of security threats, necessitating robust security measures to ensure their reliability and efficiency. This paper proposes a novel Digital Twin-Based Cyber-Physical Security Framework, incorporating AI-driven predictive maintenance and zero-trust architecture to address the evolving challenges of securing smart grids. By leveraging digital twin technology, this framework creates a real-time virtual representation of physical systems, enabling continuous monitoring and simulation for enhanced security and operational performance. Zero-trust security principles are integrated to ensure that no entity, whether inside or outside the network, is trusted by default, thus significantly reducing the risk of cyber-attacks. Additionally, AI-driven predictive maintenance enhances the framework’s reliability by proactively identifying potential failures before they occur, reducing downtime and improving system resilience. Through the development and simulation of this framework, including attack and failure scenarios, the paper demonstrates that the proposed system outperforms traditional methods in terms of anomaly detection, system downtime, and response times. The integration of predictive maintenance allows for early identification of component failures, thus enhancing the overall resilience of the grid. The zero-trust architecture further strengthens the cybersecurity posture, preventing unauthorized access and attacks. The study also identifies challenges, such as data synchronization and scalability, which must be addressed for broader implementation in large-scale smart grid systems. The findings suggest that the proposed framework could play a critical role in the future evolution of smart grid security, offering valuable insights for researchers and practitioners.  

Background: Maritime engineering has traditionally relied on reactive and preventive maintenance strategies, often leading to operational inefficiencies, unplanned downtime, and excessive costs. With the rise of smart ship technologies, predictive maintenance (PdM) has emerged as a data-driven solution, leveraging sensor-based monitoring and real-time diagnostics to optimize ship maintenance. However, its integration into maritime education remains underexplored, particularly in training vessels used for vocational learning. Original Value: This research contributes new insights into the feasibility, effectiveness, and educational relevance of predictive maintenance in maritime vocational training. Unlike previous studies that focus on commercial ship applications, this study examines PdM within the context of training vessels at Poltekpel SULUT, bridging the gap between academic training and industry expectations. Objectives: The study seeks to answer: How does predictive maintenance improve the efficiency, cost-effectiveness, and reliability of naval auxiliary systems in training vessels? Methodology: A qualitative approach was employed, integrating sensor-based performance analysis, structured interviews, and questionnaire surveys involving cadets, instructors, and industry professionals. Data were analyzed through thematic categorization, cross-group comparisons, and narrative synthesis. Results: PdM demonstrated high effectiveness in reducing downtime (92/100), optimizing maintenance efficiency (91/100), and aligning with industry practices (89/100). However, challenges in sensor accuracy (85/100) and training integration were identified. Conclusions: The findings highlight the necessity of incorporating predictive maintenance into maritime training curricula to equip future engineers with the skills required for Industry 4.0 maintenance solutions, ensuring better operational efficiency and sustainability in the maritime sector.

The maintenance of non-aviation defense equipment (main weapon system) is a critical aspect in maintaining operational readiness. However, the Maintenance, Repair, and Overhaul (MRO) system in Indonesia still faces limitations due to manual reporting, inefficiency in spare parts management, and the lack of integration of the Life Cycle Cost (LCC) approach. This study aims to design and develop the Integrated Cavalry Monitoring and Maintenance System (ICMMS) based on a web application that integrates sensors, real-time data analytics, and LCC calculation. The prototyping method was used, involving design, development, integration, and testing phases on the Maung Tactical Vehicle and Anoa Armoured Personnel Carrier at PT Pindad. The results of the prototype implementation showed a significant increase in maintenance efficiency: damage reporting time decreased from ±3 hours to ±1 minute, critical component identification became 95% faster, and maintenance scheduling shifted from reactive to predictive. Additionally, the integration of the LCC algorithm allows for more accurate maintenance cost estimation, supporting technical and strategic decision-making. This study demonstrates that ICMMS based on LCC can be an innovative digital solution to enhance MRO effectiveness and operational readiness of non-aviation defense vehicles in Indonesia. It is expected that this system will improve the resilience and cost-effectiveness of managing Indonesia’s military vehicle fleet.

This study analyzes the effectiveness of implementing predictive maintenance (PdM) on the final drive components of the Komatsu PC200-8 unit at PT. Antareja Mahada Makmur, Site PT. Multi Harapan Utama, East Kalimantan, in an effort to reduce downtime and operational losses. Before the implementation of PdM in 2022, there were 12 repair cases for the final drive with a total downtime of 772.1 hours, repair costs amounting to IDR 310.6 million, rental income loss of IDR 208.03 million, and total losses of IDR 518.63 million. In 2023, during the PdM transition phase, the number of cases decreased to 4, with a total loss of IDR 252.05 million, although downtime remained high (714.6 hours) due to the limited scope of PdM implementation on certain units and components. In 2024, with full PdM implementation, the number of repair cases decreased to 5, with total downtime of only 96 hours and losses of IDR 45.75 million. The cost of PdM implementation for the year was only IDR 21.9 million. As of July 2025, no further damage to the final drive has been recorded, demonstrating a significant improvement in equipment reliability. The reduction in total losses from 2022 to 2024 amounted to IDR 472.88 million, indicating PdM’s effectiveness in avoiding significant costs through condition monitoring methods such as oil analysis, magnetic plug rating, thermal inspection, and oil leak testing (floating seal). The findings of this study confirm that PdM is effective in reducing downtime, repair costs, and enhancing asset management in the mining sector. It also improves equipment reliability and overall operational efficiency, proving PdM to be a successful strategy in reducing losses, increasing productivity, and supporting the sustainability of company operations.

The Paiton Steam Power Plant (PLTU) is one of the main sources of electrical energy in East Java, which plays a vital role in maintaining a sustainable electricity supply. The reliability of generator units is a key element in maintaining stable energy distribution. However, the high frequency of sudden generator failures poses serious challenges, such as increased downtime and increased maintenance costs. To address these challenges, this study aims to design a generator maintenance prediction model based on the Naive Bayes algorithm with a predictive maintenance approach. This study uses historical maintenance data and key sensor parameters such as temperature, oil pressure, and vibration as input. The data is analyzed through several stages, namely data preprocessing, selection of relevant features, and labeling generator conditions into three categories: Normal, Warning, and Critical. The Naive Bayes model is trained to classify the data probabilistically to generate predictions of future generator conditions. Model evaluation using accuracy metrics and a confusion matrix shows that the model successfully achieved an accuracy rate of 89% and was able to provide early warnings of potential failures up to 3 days before failure occurs. The implementation of this system is expected to support the shift in maintenance strategies from reactive and scheduled systems to data-driven predictive systems. Implementing failure predictions allows the technical team at the Paiton PLTU to conduct planned maintenance, avoid sudden disruptions, and extend equipment lifespan. Thus, this model has the potential to reduce operational downtime by up to 25%, while providing significant savings in operational and logistics costs. This research also shows that integrating machine learning technology into energy facility management can improve the efficiency and resilience of the overall electric power system.

This The maintenance objectives of a C6240A-type lathe involve a number of crucial aspects, including gears, tool bits, toolposts, lower sled and upper sled. This study was to investigate effective maintenance strategies for each of these components, focusing on preventive and predictive maintenance. Gears: Regular lubrication is also necessary to ensure smooth gear movement and prevent excessive friction that could lead to failure. Chisel Bits: Monitoring the condition of the tool blade needs to be done regularly. Toolpost: The toolpost needs to be checked periodically to ensure availability and safety of the cutting tool. Bottom and Top Slings: Maintenance of the lower and upper slings involves checking for tension and wear. C6240A type lathe users can minimize the risk of failure, improve operational efficiency, and extend the life of the machine. The method used in this research is a qualitative method, by means of observation and interviews. The results of this study are that the gears can be more durable because lubrication is always given and not easily thirsty, the tool blade is not easily blunted because the workpiece is fed little by little, and frequent honing is done to keep it sharp. The locking toolpost is not easily damaged if you use a rubber hammer when locking, the bottom row makes a change of ashock so that it can do automatic turning. This maintenance can increase the life of the lathe for longer operation.

Predictive Maintenance is a maintenance activity that focuses on monitoring equipment conditions in real-time and analyzing data to predict potential failures before they occur, allowing repairs to be made in a timely manner before major damage occurs. One of the methods used in predictive maintenance is "Infrared Thermography” or use of technology thermal imaging technology. In the context of predictive maintenance, thermography can be used to identify problems that are not visible to the naked eye, such as poor electrical connections, excessive heat buildup, or damage to components that cause heat leakage The purpose of this study was to determine the implementation of Predictive Maintenance with Infrared Thermography method on electrical equipment at PT PHM. The method used in this research is the observation method with primary and secondary data collection. The results showed that the implementation of predictive maintenance with the Infrared Thermography method on electrical equipment and systems at PT PHM was effective in helping the company avoid unnecessary costs and improve operational efficiency. Predictive maintenance allows companies to perform maintenance to identify potential damage before it occurs and can take preventive action so as to reduce repair costs, and operational productivity.

Background: The rapid advancement of digital technologies in the Industry 4.0 era has transformed industrial mechanical systems into highly interconnected and data driven environments through the integration of sensors, the Internet of Things (IoT), data analytics, and cyber physical systems. This increasing complexity requires more adaptive and accurate monitoring and prediction methods than conventional simulation approaches, which often face limitations in capturing real time dynamic system behavior. Objective: This study aims to develop a predictive performance model for industrial mechanical systems by integrating Digital Twin technology with Physics Informed Machine Learning in order to improve monitoring accuracy and support predictive maintenance strategies. Methods: This research adopts a data driven modeling and simulation approach by developing a digital representation of an industrial mechanical system that is connected to real time sensor data. The prediction model is constructed using a Physics Informed Neural Network (PINN), which integrates operational data with physical principles governing system dynamics. The research process includes the development of a Digital Twin model, integration of sensor data, training of the PINN model, model validation using experimental data, and evaluation of prediction performance using statistical metrics. Results: The results indicate that the integration of Digital Twin technology and PINN significantly improves the prediction accuracy of industrial mechanical system performance compared with conventional simulation methods and purely data driven machine learning models. The proposed model is capable of representing system dynamics more consistently, accurately following sensor data patterns, and providing strong potential for supporting machine condition monitoring and predictive maintenance strategies in modern industrial environments.