Publication Search

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.

The interval between 2018 and 2025 represents a defining epoch in financial assurance, characterized by a systemic collision between traditional audit methodologies and the exponential sophistication of fraudulent actors. This research employs a comprehensive library research methodology, utilizing Systematic Literature Review (SLR) to evaluate the evolving landscape of audit and fraud. The study traces the theoretical migration from Cressey’s Fraud Triangle to multidimensional frameworks like the Fraud Pentagon, which emphasizes the roles of arrogance and competence. Through a forensic examination of catastrophic audit failures including Wirecard, FTX, and the emerging risks of crypto-assets, the research identifies recurring patterns of auditor failure in assessing operational risks and internal controls. Furthermore, the report analyzes the dual-edged impact of Artificial Intelligence (AI); while machine learning algorithms offer enhanced detection capabilities, the rise of Generative AI (GenAI) and deepfake technology has empowered perpetrators to execute sophisticated "synthetic reality" frauds. The study critically evaluates regulatory responses, particularly the revision of International Standard on Auditing (ISA) 240, which mandates a more proactive "fraud lens." The findings suggest that the auditing profession faces an existential crisis of relevance, necessitating a fundamental shift toward a forensic mindset supported by advanced technological integration.

This study explores the conceptual integration of Environmental, Social, and Governance (ESG) reporting and data analytics within the framework of sustainable digital accounting, with Industry 5.0 acting as a moderating paradigm. As organizations increasingly face demands for transparency, ethical governance, and sustainable operations, the limitations of traditional accounting systems have become evident. ESG reporting plays a crucial role in communicating non-financial performance and guiding strategic decisions aligned with stakeholder interests and regulatory expectations. However, the effectiveness of ESG disclosures is often hindered by fragmented data structures, inconsistent standards, and insufficient technological support. Data analytics, when integrated into ESG processes, enhances the precision, timeliness, and reliability of sustainability disclosures through predictive modeling, anomaly detection, and real time performance monitoring. Yet, despite its potential, the adoption of data analytics in accounting remains limited and under-theorized. Industry 5.0 introduces a human centric approach to technological transformation, emphasizing ethical innovation, inclusivity, and resilience. By positioning Industry 5.0 as a contextual and moderating framework, this study offers a novel perspective on how ESG and data analytics can synergize to create ethically aligned, future-ready accounting systems. Employing a systematic literature review, the research develops a conceptual model linking ESG, data analytics, and Industry 5.0, providing insights for academics, practitioners, and policymakers aiming to embed sustainability into digital accounting systems. The findings underscore the importance of aligning digital tools with ethical and societal values to advance accountable and sustainable business practices..    

Food safety is crucial for maintaining public health, especially in rural areas that face limitations in knowledge and resources to detect harmful substances in food products. The PKK Group of Borisallo Village, Gowa Regency, has great potential to become an agent of food safety education and monitoring. However, they still face limitations in understanding the impact of pathogenic microbes, pesticide residues, and harmful chemicals such as formalin, borax, and synthetic dyes. This Community Service Program (PKM) aims to enhance the knowledge and skills of PKK members through education and training on detecting harmful substances using simple organoleptic methods and household tests. The activities include counseling, self-detection demonstrations, and providing supporting tools such as portable stoves and frying pans for home practice. The results show a significant improvement in participants' knowledge, especially regarding the characteristics of contaminated food and natural inspection techniques. The PKK group also showed high enthusiasm in disseminating food safety information. This program enhances the capacity of PKK as agents of change, strengthens the culture of food safety at the household level, and supports sustainable education through collaboration with the village government and health centers. It is hoped that this can be replicated in other villages.

Food safety is crucial for public health, particularly in rural areas with limited access to information and food additive (BTP) detection tools. The PKK group in Borisallo Village faces challenges due to low awareness about the dangers of BTP such as formalin, borax, Rhodamin B, methanol yellow, and synthetic sweeteners, as well as limited skills in detecting these substances. This Community Service (PKM) program aims to enhance the knowledge and skills of PKK members through education and training on natural BTP detection and food processing. Methods include counseling, demonstrations using organoleptic and household tests, and practicing with natural alternatives for preservatives, sweeteners, and food colorings. Results showed a significant increase in participants' understanding and detection ability, as seen in the pretest and posttest evaluations. The provision of portable stoves and pans also ensures sustainability of the activities. This program strengthens the PKK's role in food safety supervision at the village level and suggests further training, additional detection tools, and more collaboration with health centers and local governments to sustain the initiative.

Diabetes mellitus (DM) is a metabolic disease characterized by elevated blood glucose levels. This condition includes several types, such as type 1 and type 2 diabetes. It occurs when the body is unable to produce enough insulin, experiences impaired insulin function, or faces a combination of both. As a result, glucose cannot be effectively absorbed by the body’s cells and accumulates in the bloodstream, leading to high blood sugar or hyperglycemia. If this condition persists, the buildup of glucose can cause various disorders in different organs. Without proper management, diabetes may lead to severe and life-threatening complications. Chronic hyperglycemia in diabetic patients is associated with long-term damage, dysfunction, and even failure of vital organs such as the eyes, kidneys, nerves, heart, and blood vessels. In this activity, percentage-based methods and demonstrations of blood glucose testing were used as educational tools for students. The results showed that students’ awareness of the dangers of elevated blood sugar levels remains low. Many of them are still unfamiliar with the importance of regular blood glucose monitoring and maintaining a healthy lifestyle to prevent diabetes. Therefore, continuous efforts are needed to provide education and outreach on the prevention and early management of diabetes mellitus.

Information system security faces serious challenges due to increasingly complex cyber attacks. Intrusion Detection Systems (IDS) require efficient approaches to handle high-dimensional data such as the NSL-KDD dataset with 41 features. This study aims to implement the Genetic Algorithm (GA) for feature selection on the NSL-KDD dataset to improve the efficiency and accuracy of network attack detection. The method used is computational experimental research, involving data preprocessing, GA implementation for feature selection, building a classification model using Random Forest, and performance evaluation based on accuracy, precision, recall, F1-score, and computation time. The results show that GA successfully reduced features from 41 to 12 features (70.7% reduction), significantly improving computational efficiency. However, model accuracy slightly decreased from 0.4973 to 0.4951, indicating that while GA is effective for feature selection, the elimination of certain features may reduce classification capability. The implication of this study is that GA can be used as a tool to simplify intrusion detection models, but it should be combined with parameter optimization and data imbalance handling to achieve more optimal performance.  

Law enforcement against narcotics crimes in Indonesia faces various complex problems, both from legal aspects, institutional structure, and community legal culture. Even though Law Number 35 of 2009 concerning Narcotics has been enacted, the circulation and abuse of narcotics is still increasing every year. This study aims to analyze problems in law enforcement against perpetrators of narcotics crimes and identify the factors that affect them. The research method used is normative juridical with a legislative approach and a case approach. The results of the study show that the main obstacles lie in the lack of coordination between law enforcement agencies, weak integrity of the apparatus, overlapping regulations, and low public legal awareness. In addition, the limitation of supporting facilities and infrastructure such as rehabilitation facilities, forensic laboratories, and early detection technology also hinders the effectiveness of case handling. The high modus operandi of increasingly sophisticated narcotics networks is also a challenge for the authorities. Therefore, strengthening regulations, increasing the capacity of human resources, utilizing modern technology, and more intensive collaboration between agencies need to be optimized as a comprehensive effort to strengthen the effectiveness of law enforcement in the narcotics sector.

The rapid growth of web technologies and online services has increased the exposure of web applications to cyber threats such as Cross-Site Scripting (XSS) and SQL Injection (SQLi). Conventional rule-based mechanisms, such as Web Application Firewalls (WAFs), often fail to detect emerging attack patterns. To address this, Machine Learning (ML) and Deep Learning (DL) have emerged as adaptive approaches for enhancing web attack detection. This study performs a Systematic Literature Review (SLR) following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines to analyze recent ML/DL-based detection methods. Of the 263 retrieved studies, 15 met the inclusion criteria for detailed review. The findings reveal that Random Forest (RF), Support Vector Machine (SVM), Convolutional Neural Network (CNN), and Long Short-Term Memory (LSTM) are the most applied algorithms. At the same time, recent works emphasize Transformer-based and hybrid ML–DL models. These approaches achieved robust performance (accuracy 85–97%, F1-score >90%) but still face challenges in dataset representativeness, class imbalance, and computational cost. This review highlights future research directions in Explainable Artificial Intelligence (XAI), Federated Learning (FL), and adversarial robustness to develop more efficient and trustworthy web attack detection systems.

Application of Multi-Layer Perceptron neural network to fault classification in high-voltage transmission lines is demonstrated in this paper. Different fault types on protected transmission line should be detected and classified rapidly and correctly. This paper presents the use of Discrete Wavelet Transform energy features combined with zero sequence current magnitude as input features for neural network classifier. The proposed method uses eight extracted features to learn hidden relationship in fault signal patterns. Using proposed approach, fault detection and classification of all 11 fault types could be achieved with high accuracy. Improved performance is experienced once the neural network is trained sufficiently with 1188 fault samples, thus performing correctly when faced with different system conditions. Results of performance studies show that proposed neural network-based classifier achieves 96.18% average accuracy, which demonstrates that it can improve the performance of conventional fault classification algorithms, which in turn can provide more efficient solutions in the management and protection of high voltage electrical systems.

The background of this research is based on Cervical cancer, a disease that poses a serious threat to women's health in Indonesia. Early detection through prevention programs such as IVA (Visual Inspection with Acetic Acid) is a crucial step in reducing the number of cervical cancer cases. In this effort, an effective communication strategy from the Pasar Minggu District Health Center plays a very important role. This study focuses on the prevention program and communication strategies implemented at the Pasar Minggu District Health Center. This study uses a qualitative approach with a descriptive study method. Data collection techniques were carried out through in-depth interviews with Health Center officers, observations of socialization activities, and documentation. The results show that the communication strategies implemented by the Pasar Minggu District Health Center include informative, educational, and persuasive communication strategies. The Health Center utilizes various communication channels such as social media and direct counseling. Obstacles faced include low levels of public health literacy, stigma against IVA examinations, and limited human resources.

Millions of people suffer from malaria, one of the most serious parasitic diseases that threatens human life and causes high rates of morbidity and mortality, particularly in tropical and subtropical regions. Traditional diagnostic methods, such as blood smear examination, which can be performed using a microscope, face many challenges due to the inaccuracy of manual analysis and the reliance on individual skills. Therefore, the use of machine learning or deep learning algorithms to automate malaria detection offers promising solutions to improve accuracy, reduce diagnosis time, and enhance scalability. In this paper, a multi-class convolutional neural network (CNN)-based model is designed to classify cells infected with Plasmodium falciparum (P. falciparum) and Plasmodium vivax (P. vivax) and uninfected cells from blood smears, as most severe cases and deaths are caused by P. falciparum and P. vivax. This is achieved by building and training a CNN from scratch, rather than using transfer learning from pre-trained models. The proposed network was trained and tested on the Kaggle dataset, which consists of 27,558 images of infected and uninfected individuals. These images were divided into 13,779 images of uninfected individuals, 6,890 images of individuals with P. falciparum malaria, and 6,889 images of individuals with P. vivax malaria. The images were preprocessed using several operations, including blurring, denoising, and morphological processing. The proposed model achieved the best evaluation accuracy when compared with other deep learning algorithms, with an accuracy rate of 96.5%, a sensitivity rate of 95%, a specificity rate of 97.6%, and an F1-score rate of 96.5%. These results demonstrate the effectiveness of the proposed model as a tool to assist clinicians in malaria diagnosis, reducing reliance on manual analysis.

Tomato crop yields face significant threats from plant diseases, with existing deep learning solutions often computationally prohibitive for resource-constrained agricultural settings; to address this gap, we propose Efficient Disease Attention Network (EDANet), a novel lightweight architecture combining depthwise separable convolutions with hybrid attention mechanisms for efficient Tomato disease recognition. Our approach integrates channel and spatial attention within hierarchical blocks to prioritize symptomatic regions while utilizing depthwise decomposition to reduce parameters to only 104,043 (multiple times smaller than MobileNet and EfficientNet). Evaluated on ten tomato disease classes from PlantVillage, EDANet achieves 97.32% accuracy and exceptional (~1.00) micro-AUC, with perfect recognition of Mosaic virus (100% F1-score) and robust performance on challenging cases like Early blight (93.2% F1) and Target Spot (93.6% F1). The architecture processes 128×128 RGB images in ~23ms on standard CPUs, enabling real-time field diagnostics without GPU dependencies. This work bridges laboratory AI and practical farm deployment by optimizing the accuracy-efficiency tradeoff, providing farmers with an accessible tool for early disease intervention in resource-limited environments.

Illegal streaming sites are one of the forms of copyright infringement that are rampant in the digital era and have a significant impact on the creative industry, the economy, and law enforcement. This research analyzes the Indonesian legal framework consisting of Law No. 28 of 2014 on Copyright, Law No. 11 of 2008 jo. Law No. 19 of 2016 on Electronic Information and Transactions, as well as the technical regulations of the Ministry of Communication and Information (Kominfo) in addressing these violations. The findings indicate that although the legal framework is in place, enforcement still faces technical, legal, social, and complex cross-border challenges. These sites often utilize foreign servers, mirror domains, and anti-blocking technology to evade blocking, thus requiring international cooperation thru mechanisms such as Mutual Legal Assistance (MLA) and coordination with the World Intellectual Property Organization (WIPO). An effective counter-strategy must be multidimensional, encompassing regulatory strengthening, the use of detection technologies such as digital watermarking and content ID systems, as well as public education to curb the demand for illegal content. A comparison with the United States, Japan, and South Korea shows that proactive enforcement, a quick notice-and-takedown mechanism, and industry cooperation have proven effective in reducing violations. In conclusion, the eradication of illegal streaming sites in Indonesia requires continuous synergy between the government, industry, and society to protect copyright and the sustainability of the creative industry ecosystem in the digital era.

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 DH Building of the Electrical Engineering Study Program at Udayana University faces significant challenges in energy efficiency, as it still relies on conventional electrical systems. User negligence, such as forgetting to switch off lights and air conditioners (AC) after use, often results in unnecessary energy waste and increased operational costs. This issue highlights the urgent need for smart solutions capable of automating energy management, reducing waste caused by human error, and supporting the creation of a more efficient and sustainable campus environment. To address this problem, this study designs and implements a smart building system based on the Internet of Things (IoT). The system employs a NodeMCU ESP32 microcontroller as the main processing unit, integrated with a series of sensors including a DHT22 sensor for monitoring temperature and humidity, an MQ2 sensor for smoke detection, a PIR sensor for motion detection, and a PZEM-004T sensor for monitoring energy consumption. Control of electronic devices such as lights and AC units is carried out both automatically and manually through relay modules connected to the system. All sensor data and control functions are accessed via a web interface developed using the Laravel framework and a MySQL database. The testing results indicate that the designed system was successfully implemented and functions as expected. Sensor testing demonstrated high accuracy compared to standard measuring instruments, while the electronic device control system achieved an average response time of approximately 3.6 seconds, proving its reliability. Overall, the system provides a comprehensive solution for energy consumption monitoring and control, while also enhancing comfort and safety in the DH Building, in line with the goals of energy efficiency and facility modernization.

Belawan Port is a strategic port in North Sumatra with export-import activities reaching 5,000 tons per year, making it an important terminal in supporting international trade and regional economic growth. This research examines the role of daily work reports in dry bulk cargo unloading operations at PT. Wahana Intradermaga Niaga Belawan as a Stevedoring Company (PBM) responsible for ensuring the smooth process of loading and unloading at the port. The research methodology employs library research approach and direct field observation. The dry bulk cargo unloading process involves the inaportnet system and operates for 24 hours with three work shifts. Activities include four main types of operations: stevedoring (transferring cargo from ship to wharf), cargodoring (transfer from wharf to warehouse), delivery (shipment outside the port), and receiving (acceptance from factory to warehouse). The daily report document is a list of all cargo unloading activities during 24 hours at wharf 112. Daily work reports function as structured documentation of daily activities, conveying work progress updates, supporting performance monitoring, and serving as a reference for operational evaluation. The report's usefulness includes strengthening accountability, reducing misunderstandings, early problem detection, consistency in task implementation, and orderly documentation of unloading activities. Operational preparation involves permit processing according to port regulations, equipment preparation such as hopper, conveyor belt, grab, excavator, wheel loader, and sling ropes. Supporting documents include Bill of Lading, Cargo List, Cargo Manifest, and various operational reports. Obstacles faced include the influence of bad weather, human resource constraints related to workforce professionalism, and land transportation barriers. Related institutions include cargo owners, PBM, shipping agents, port authorities, PT. Pelindo Belawan, and EMKL companies that coordinate to ensure smooth dry bulk cargo unloading operations.

In the automotive manufacturing industry, efficiency in quality control is a crucial factor to ensure consistent product quality. Conventional Quality Assurance (QA) processes using manual record-keeping often face challenges such as delayed reporting, human errors, and difficulty in tracking historical data. This study aims to design and implement a QA performance dashboard based on digital forms at PT Dharma Polimetal, Tbk, to enhance efficiency in production quality control. The research methodology includes direct field observation, collection of production and QA data, mapping of QA process flows, interactive dashboard interface design, and system trial implementation. The designed dashboard focuses on four main aspects: QA Incoming, QC Line, QC Gate, and Customer Handling, each containing measurable performance indicators and quality parameters. Initial implementation results indicate significant improvements in QA process monitoring, faster reporting of inspection results, and easier real-time data access for both production teams and management. The system enables early detection of potential quality issues, supports rapid decision-making, and facilitates internal and external audits. Moreover, the use of digital forms within the dashboard enhances data accuracy, minimizes human error, and creates structured historical records for long-term analysis. This study provides a tangible contribution to the digitalization of QA systems, strengthening sustainable quality control practices in the automotive industry, thereby ensuring consistent productivity and product quality.

Attendance is an essential activity in both educational institutions and companies, serving as an indicator of discipline, presence, and individual responsibility. Conventional attendance systems that still rely on manual journals often face several problems, such as vulnerability to manipulation, data loss, and physical damage. Meanwhile, modern methods such as fingerprint, QR code, RFID, and GPS are not entirely ideal since each has its own limitations in terms of cost, accuracy, user convenience, and potential misuse. For instance, fingerprint systems raise hygiene concerns due to shared use, while QR code and GPS methods are prone to fraud and location spoofing. To address these challenges, this study proposes a face-based attendance simulation system by integrating the YOLOv8 algorithm for face detection and Local Binary Pattern Histogram (LBPH) for face recognition. YOLOv8 was chosen for its ability to detect faces in real time with high speed and accuracy, while LBPH is employed for face recognition due to its robustness in handling variations in facial features and its relatively low computational requirements. This makes the system efficient even when implemented on medium-specification devices. The system was tested on 25 participants with a total of 250 attendance attempts. Based on the confusion matrix analysis, the system achieved outstanding performance with 98.4% accuracy, 98.4% precision, 100% recall, and a 99.2% F1-score. Furthermore, the system automatically recorded attendance dates and times with an average latency of 69.185 ms, proving its capability to operate quickly and reliably in real-world scenarios. Nevertheless, several limitations were observed, such as decreased accuracy when the face moved too quickly during image capture, as well as potential performance degradation under extreme lighting conditions. Despite these challenges, the proposed system demonstrates excellent performance and offers a promising solution for efficient, hygienic, and fraud-resistant attendance management applicable to both educational and professional environments.

The spice industry faces significant challenges in maintaining product weight consistency as part of quality assurance and compliance with production standards. A case at PT X revealed that a newly installed filling machine produced deviations from the target weight of 50 grams, with hypothesis testing showing that out of 30 samples, 17 samples fell outside the  confidence interval. To mitigate this issue, this study proposes the development of a real-time data-driven Decision Support sistem (DSS) combined with statistical approaches. The methodology includes two-tailed hypothesis testing to detect weight deviations and Failure Mode and Effects Analysis (FMEA) to identify dominant failure causes based on high Risk Priority Numbers (RPN), such as delayed machine calibration, operator error, and worn-out machine components. These findings serve as the foundation for designing the DSS architecture, which consists of sensor input modules, statistical data processing, risk mapping, and an automated corrective recommendation engine. The sistem is designed to enable early detection of deviations, accelerate response time to quality issues, and support data-driven decision-making on the production floor. The study concludes that a structured implementation of DSS can be an effective strategy to improve product weight consistency and enhance operational efficiency in spice manufacturing.