Publication Search

This study aims to compare the performance of three boosting algorithms, namely XGBoost, LightGBM, and CatBoost, to predict the success of MSMEs. The data used consists of 250 entries with 13 attributes that include business actor characteristics, initial capital, industry experience, financial record-keeping, internet utilization, business planning, partnerships, and the target variable success. The pre-processing stage includes checking for missing values, standardizing numerical attributes, and splitting the data into 80% training data and 20% test data. The evaluation results show that XGBoost provides the best performance with an accuracy of 0.92, precision of 0.8333, recall of 0.8333, F1-score of 0.8333, and ROC-AUC of 0.9715. LightGBM has an accuracy of 0.88, while CatBoost achieves an accuracy of 0.90. The research results show that XGBoost has the best ability to classify successful and unsuccessful MSMEs. The feature importance results also show that the success of MSMEs is influenced by a combination of several key factors. This research emphasizes that boosting algorithms are effectively used as predictive models to support the analysis of MSME success.

Tomato (Solanum lycopersicum) is a high-value horticultural commodity in Indonesia, yet its cultivation is frequently disrupted by leaf diseases that are difficult to distinguish visually. Diseases such as Bacterial Spot, Early Blight, and Tomato Yellow Leaf Curl Virus often present overlapping visual symptoms, making early and accurate diagnosis a significant challenge for farmers. The manual identification methods currently in use are inefficient and error-prone, ultimately leading to reduced crop yield  and quality. The general objective of this study is to develop software capable of automatically classifying tomato  leaf diseases. Specifically, this research aims to implement the MobileNetV3 Small architecture based on Convolutional Neural  Network (CNN) with ImageNet pre-trained weights to classify 10 types of tomato leaf diseases. The research methodology encompasses dataset collection from Kaggle comprising 10,000 images (1,000 per class), image pre-processing through resizing to 224x224 pixels, and normalization, as well as hyperparameter optimization (optimizer, learning rate, epoch, batch size) via scheduler. Model performance is evaluated using a confusion matrix encompassing accuracy, precision, recall, and F1-score.

Abstract. PT XYZ, a wooden furniture manufacturing company, served as the research site for this study which applied the Theory of Constraints (TOC) method to analyze production performance and identify bottlenecks. The company faces capacity imbalances between workstations, resulting in production targets that have not been achieved optimally. Data collection involved direct observation and interviews with related parties in the production area. The analysis was conducted by comparing the required capacity with the available capacity at each production workstation. The findings reveal that solid processing, machining, sanding, assembling, painting, and packing have sufficient available capacities to meet production requirements, thus categorized as non-bottleneck processes. In contrast, the panel processing station is identified as the main bottleneck due to its highest workload among all processes. By implementing the Theory of Constraints, the company can identify major constraints and establish improvement priorities to enhance production flow smoothness. It is expected that improvements in bottleneck processes will increase production efficiency, balance capacity among workstations, and support more optimal achievement of production targets. Keywords: bottleneck; capacity; manufacturing; production performance; theory of constraints   Abstrak. PT XYZ sebuah perusahaan manufaktur furnitur kayu, menjadi lokasi penelitian ini yang menggunakan metode Theory of Constraints (TOC) untuk menganalisis kinerja produksi dan mengidentifikasi bottleneck. Perusahaan menghadapi ketidakseimbangan kapasitas antar stasiun kerja yang menyebabkan target produksi belum terdengar secara optimal. Pengumpulan data meliputi observasi langsung dan wawancara dengan pihak terkait di area produksi. Analisis dilaksanakan dengan membandingkan kapasitas yang dibutuhkan terhadap kapasitas yang tersedia pada setiap stasiun kerja produksi. Hasil penelitian menunjukkan bahwa proses pembahanan solid, machining, sanding, assembling, painting, dan packing memiliki kapasitas yang tersedia yang masih mampu memenuhi kebutuhan produksi, sehingga termasuk kategori non-bottleneck. Sebaliknya, stasiun kerja pembahanan panel diidentifikasi sebagai bottleneck utama karena memiliki tingkat beban kerja tertinggi di antara seluruh proses. Dengan penerapan Theory of Constraints, perusahaan dapat mengidentifikasi kendala utama dan menentukan prioritas perbaikan untuk meningkatkan kelancaran aliran produksi. Diharapkan perbaikan pada proses bottleneck dapat meningkatkan efisiensi produksi, menyeimbangkan kapasitas antar stasiun kerja, serta mendukung pencapaian target output perusahaan secara lebih optimal. Kata kunci: bottleneck; kapasitas; kinerja produksi; manufaktur; theory of constraints

Plastic waste, particularly plastic bottles, has become one of the major challenges in urban environmental management due to its increasing volume and potential to pollute ecosystems. In Bogor City, plastic waste constitutes a significant proportion of daily municipal solid waste, highlighting the need for effective and sustainable waste management models. This study aims to analyze the role of the Reduce, Reuse, Recycle Waste Processing Facility (TPS3R) in managing plastic bottle waste in Bogor City, with a case study of the Eco Techno Park at Ibn Khaldun University (UIKA) Bogor. The research employed a descriptive qualitative method using a case study approach. Data were collected through field observations, in-depth interviews with facility managers, and reviews of relevant documents and literature. The findings indicate that the TPS3R Eco Techno Park has successfully implemented the 3R principles through an integrated system supported by environmentally friendly technologies and a circular economy framework. Plastic bottle waste is managed through sorting, shredding, and recycling processes to produce value-added products, including plastic pellets, handicrafts, and construction materials. In addition, organic waste management is integrated through the cultivation of Black Soldier Fly (BSF) larvae. The facility has contributed to reducing the volume of waste sent to landfills by approximately 18%. The study implies that strengthening regulatory support, enhancing community participation, and developing circular economy-based business models are essential to ensure the long-term sustainability of waste management programs and support Bogor City's waste reduction targets.

The benefit of cardboard waste as a headpiece is to encourage the entrepreneurial spirit to continue to be creative in processing this inorganic waste into products at prices that are relatively cheap compared to prices on the market. Data collection uses the methods of Observation, Literature, Experimentation, Documentation, Interviews, Questionnaires. The author carried out the process of using cardboard waste to make headpieces through several stages, not only that, the author also carried out experiments 3 times using different dyes. In experiment 1 using gold pilok, experiment 2 used gold pilok then sprinkled with gold glitter, then experiment 3 used wall paint coloring sprinkled with gold glitter. The author conducted a sensory test regarding the public's acceptance of headpiece products made from cardboard waste. The highest average results were obtained in experiment 3 with a total average color of 2.87 which means quite appropriate, a total average texture of 2.63 which means quite suitable, a total average design of 1.77 which is less appropriate, while the total average ease of use is 2.67, which means it is quite appropriate. So the results obtained were that as many as 30 respondents preferred product 3 with quite appropriate criteria. From the results of the products made, the coloring still needs to be improved so that they are more perfect and the designs are developed to be more varied so that they can be used as a reference for further research.

The implementation of web-based archive digitalization can help improve performance in the field of archive recording, which was previously still manual at the Naval Academy. Some of the obstacles include archiving that is still manual, many files lost due to piled-up storage systems, and human resources in the field of archiving that are still lacking. This study aims to increase time and cost efficiency in archive processing, reduce the risk of loss and damage to archives, and improve the abilities and skills of human resources in the process of using digital archives. This type of research is descriptive qualitative. The data collection techniques are interviews, observation, and documentation with respondents as well as appropriate documentation. The data analysis was carried out through the processes of data reduction, data display, as well as verification or drawing conclusions. The research results can help improve the efficiency and security of archiving by minimizing risks, so that archive management can enhance efficiency and effectiveness, focusing on the implementation of web-based applications.


This study aims to formulate an engineering strategy for water quality management and pollution control in the Tiku Sub-watershed, North Musi Rawas Regency. The study used a descriptive approach by integrating selected technical data from a dissertation-based assessment with semi-structured interviews involving 20 key informants representing technical agencies, village and subdistrict governments, community leaders, artisanal and small-scale gold mining actors, farmers, and riverbank communities. The analysis focused on water quality status, water availability and demand, pollution sources, perceived impacts, implementation constraints, and priority interventions. The results showed that the Tiku Sub-watershed is under significant environmental pressure. The average pollution index was 6.65, indicating a moderately polluted status, while mercury, cadmium, phosphate, and ammonia were the dominant parameters of concern. Surface water availability remained relatively adequate at 45,842,699.79 m3/year, and the average Criticality Ratio was 0.266; however, the water pollution carrying capacity was poor. Interview results indicated that mercury use and processing waste from artisanal gold mining were the most urgent issue (35%), followed by land-cover change and riparian degradation. The recommended strategy combines mercury-free processing technology, alternative livelihood development, cross-sectoral supervision, riparian rehabilitation, and transparent water-quality monitoring. The findings imply that watershed engineering should integrate technical, institutional, and socio-economic interventions.

Embroidery is one of Indonesia’s traditional craft arts that has developed over time and continues to attract people from various social backgrounds. Embroidery techniques are commonly applied to fashion products as well as household items. The diversity of embroidery motifs and techniques provides aesthetic value and uniqueness to each product, making embroidery an important element in the development of the fashion industry. This opportunity can encourage students of the Fashion Design Program at AKS Ibu Kartini to develop their entrepreneurial potential through creativity in manual embroidery. This study employed a quantitative approach using multiple linear regression analysis as the data processing technique to determine the influence of internal factors on the entrepreneurial motivation of Fashion Design students at AKS Ibu Kartini. The research sample consisted of 65 respondents. The findings revealed that, simultaneously, the four independent variables had a significant effect on entrepreneurial motivation, with a coefficient of determination (R²) value of 0.910. This indicates that 91% of the variation in entrepreneurial motivation can be explained by technical knowledge (X1), technical skills (X2), creativity and innovation (X3), and entrepreneurial mindset (X4), while the remaining 9% is influenced by other factors outside this study. Partially, variables X1, X2, and X4 showed a positive and significant effect on entrepreneurial motivation, whereas X3 demonstrated a negative and significant effect. This finding indicates that creativity without market orientation may reduce students’ motivation to engage in entrepreneurship. The implications of this study emphasize the importance of improving students’ knowledge, skills, and entrepreneurial mindset. Furthermore, proper guidance is needed to ensure that students’ creativity and innovation are aligned with market demands in order to optimally enhance entrepreneurial motivation.

Flooding remains a recurrent hazard in the Special Capital Region of Jakarta (DKI Jakarta), causing substantial disruptions across multiple dimensions community life, including social well-being and economic activities. The availability of flood-related information through the Jakarta One Data Portal (Satu Data Jakarta) provides significant opportunities for broader data utilization; however, transforming such information into meaningful assessments regional vulnerability requires a systematic analytical approach to generate more comprehensive understanding of the conditions of individual urban villages (kelurahan). This study focuses on the classification areas based on the characteristics and impacts of flood events occurring throughout DKI Jakarta. The analysis utilizes flood event records from 2023 to 2025 and incorporates several indicators, including the number of flood occurrences, the number of affected neighborhood associations (RW), the number affected households, the number of affected residents, the number of evacuees, the number of evacuation sites, and floodwater depth. Data processing was conducted using the Knowledge Discovery in Databases (KDD) framework, encompassing data selection, cleaning and preprocessing, transformation, pattern exploration, result evaluation, and knowledge extraction. The findings demonstrate that three-cluster solution effectively captures variations in flood vulnerability levels, corresponding to low-, moderate-, and high-risk categories. A total of 96 urban villages were classified as low-risk, 27 as moderate-risk, and 46 as high-risk areas. The resulting clustering patterns provide a clearer spatial representation of flood-risk distribution across urban villages, thereby offering valuable insights for the development more targeted mitigation strategies, the prioritization of flood management interventions, and the enhancement of evidence-based decision-making processes in DKI Jakarta.

The area and distance between buildings at POLINELA Campus make the presence of adequate connector roads an important necessity to support the mobility of the academic community. This study aims to optimize the scheduling of flexible pavement construction for the POLINELA Connector Road using the Critical Path Method (CPM) based on Microsoft Project 2021. The research method used is descriptive quantitative by processing the Budget Plan (RAB) data into a project schedule consisting of 11 main activities. CPM analysis is used to determine the sequence of work, dependencies between activities, the critical path, and project completion time. The results show that the total project duration is 23 calendar days (May 11–June 6, 2026). CPM analysis identified 10 activities on the critical path with zero float, while embankment work is a non-critical activity with time flexibility. Applying a Start-to-Start (SS) relationship with a 3-day lag between excavation and embankment work allows parallel implementation, reducing the project duration by 4 days (14.8%) compared to the conventional Finish-to-Start (FS) scenario. These results indicate that implementing CPM through Microsoft Project 2021 can increase execution time efficiency, simplify schedule control, and support decision-making in road construction project management.

The rapid development of information technology drives various business sectors, including the beauty and health industry, to adopt digital systems in their operational management. A Beauty Clinic, a medium-scale aesthetic clinic, faces significant operational challenges in managing human resources, finance, and digital marketing in an integrated manner. This study aims to analyze the existing manual operational system of A Beauty Clinic using the PIECES framework (Performance, Information, Economy, Control, Efficiency, Service), design and configure an Odoo ERP Community Edition implementation integrating the HR, Invoicing/Point of Sale, and Social Marketing modules, and evaluate the operational efficiency improvements achieved. The research employed a qualitative descriptive approach with a case study method and Rapid Application Development (RAD) as the system development methodology. Data was collected through direct observation and semi-structured interviews. The results show that the manual system experienced critical problems across all six PIECES dimensions. Post-implementation evaluation indicates an estimated 80% reduction in the secretary's administrative workload (from 20–25 hours to 3–5 hours per month), over 90% reduction in payroll error frequency, and an 87% reduction in transaction processing time per cashier transaction. The implementation of role-based access control and complete audit trails fundamentally addressed the security vulnerabilities identified in the as-is analysis. This study demonstrates that Odoo Community Edition is an effective, comprehensive, and cost-efficient ERP solution for small and medium-scale beauty clinics seeking to transform from fragmented manual systems to integrated, data-driven information systems.

Document-level sentiment analysis assigns a single polarity label to an entire review, often obscuring opinion diversity within multi-sentence submissions. This limitation is particularly evident in reviews of multi-service platforms, where users frequently express heterogeneous opinions toward different aspects of the platform in the same review. To address this challenge, this study proposes a sentence-level sentiment analysis framework for Indonesian Gojek app reviews collected from the Google Play Store. The proposed framework introduces a two-stage segmentation strategy that combines punctuation-aware rules with conjunction-aware splitting based on coordinating and adversative conjunctions (e.g., tapi [but], padahal [even though]) to identify opinion boundaries and decompose mixed-sentiment reviews into independently classifiable sentence units. A total of 14,730 raw reviews collected between May and July 2025 were subjected to data cleaning and quality filtering, resulting in 7,187 valid reviews that were further segmented into 14,187 sentence-level instances. Each instance was manually annotated by three annotators using a four-class labeling scheme consisting of app-positive, app-negative, app-neutral, and service categories. Sentiment-level inter-annotator agreement, computed on the subset of instances unanimously categorized as app-related by all three annotators (n = 4,384), achieved substantial agreement (Fleiss'  = 0.636). Hyperparameter optimization was conducted using Optuna with the Tree-structured Parzen Estimator (TPE) sampler across four experimental scenarios. The best performance was achieved by IndoBERTweet under Stratified K-Fold evaluation, attaining an accuracy of 0.751 and a macro F1-score of 0.729, outperforming all IndoBERT configurations. The results demonstrate the effectiveness of domain-adaptive pre-training on informal Indonesian text and highlight the value of conjunction-aware segmentation for preserving fine-grained opinion structures in mixed-sentiment reviews. These findings suggest that domain-aligned language representations provide a practical and effective solution for sentence-level sentiment analysis of Indonesian app reviews.

The rubber processing manufacturing company PT Darmasindo Intikaret Tebing Tinggi is encountering more and more intense competition. So, managing human resources, the company's most valuable asset, is a top priority alongside meeting financial and operational goals. In human resource management, employee happiness is a key performance measure. Leadership style, discipline, and K3 are all variables that this study hopes to identify and examine in some detail. This research falls under the category of development studies and employs a quantitative methodology. This research makes use of both primary and secondary sources for its data. Findings from both primary and secondary sources are used into this research. People who work in production at PT. Darmasindo Intikaret were the subjects of this research. According to the results of the complete sampling method, the milling production staff consists of 34 individuals. In this study, the entire population is employed as a sample. The data analysis techniques utilised in this study included reliability and validity tests, tests for classical assumptions like normality and multicollinearity and heteroscedasticity, and tests for multiple regression, determination coefficients, and hypotheses like the F and T tests. Leadership style significantly and positively affects employee job satisfaction, according to the study's results. There is a favourable and statistically significant relationship between discipline and employee job satisfaction, and K3 is no exception. Employees' happiness on the job is influenced by a number of factors, including leadership style, discipline, and K3. The significance of enhancing the quality of leadership style, discipline, and 3 in boosting employee job satisfaction in a firm, particularly manufacturing companies, can be demonstrated by doing this study.

The Jambi Class II Climatological Station is the agency responsible for providing climate data and information to the public and relevant agencies. In practice, the process of handling data requests is still carried out conventionally, resulting in relatively long processing times for document management, archiving, and user service. This situation has led to several challenges, such as inefficient administrative processes, a high risk of document loss, and limitations in monitoring the status of data requests. This study aims to design a web-based Data Service Information System that can support the digitalisation of climate data services at the Class II Jambi Climatology Station. The system development method used is the prototype method, as it allows the development process to be carried out in stages based on user needs. The system design was carried out using the Unified Modelling Language (UML) as a modelling tool to describe system requirements, process flows, and user interactions with the system. The design results show that the developed system is capable of facilitating the online submission of data requests, tracking the status of requests, and managing the service database in a more structured manner. The system also provides a digital archiving feature that simplifies the management and retrieval of data request documents. The implementation of this system is expected to improve the efficiency of service processes, speed up data request processing times, enhance service transparency, and support improvements in the quality of public services at the Class IV Jambi Climatological Station.

Coal hauling activities are an important part of mining operation because they affect production continuity, cycle time efficiency, and operational safety. This study aims to analyze the requirements of road support equipment on the coal hauling road from Sector 4 to the new Coal Processing Plant (CPP) at PT. Asmin Bara Bronang, Central Kalimantan. Based on road geometry, traffic density, California Bearing Ratio (CBR), and Unsurfaced Road Condition Index (URCI). The research method used was applied research with a quantitative approach. Primary data ware collected through field measurements consisting of road geometri observations, traffic density observations, Dynamic Cone Penetrometer (DCP) testing to obtain CBR values, and road surface condition assessment using the URCI method. Secondary data were obtained from the company records. The results showed that the hauling road has a total length of 9.1 km with an average width of 16 m, and grade values ranging from -7.68% to 10.52%, which are still below the maximum standard of 12%. Traffic density reached 184 dump trucks/day, for coal hauling and 62 units/day for construction material transportation, indicating high traffic intensity. CBR values ranged from 7% to 100%, showing variations in subgrade bearing capacity. The URCI value ranged from 72,50 to 91.00, indicating fair to good road conditions. Based on the analysis of road conditions and maintenance area requirements, the recommended support equipment for maintaining the hauling road consists of 1 motor grader unit, 1 compactor unit, 1 bulldozer unit, and 1 water truck unit.

The development of social media, especially TWITTER, has become one of the main means for people to express opinions and criticism on various issues, including the performance of law in Indonesia. This study aims to analyze public sentiment towards the performance of law based on TWITTER user comments using the Naïve Bayes algorithm. The research data consists of 1004 comments collected from several videos related to legal topics. The analysis process includes the stages of data crawling, pre- processing (text cleaning, normalization, and tokenization), labeling sentiment into positive, negative, and neutral, and testing the Naïve Bayes model. The results show that the Naïve Bayes algorithm is able to classify sentiment with an accuracy level of 93.73%. The distribution of sentiment from 1004 comments shows that the majority of public opinion is (negative/positive/neutral), which indicates that public perception of the performance of law is still (critical/positive). These findings are expected to be input for related parties to understand public opinion and improve the quality of legal performance in

This research is motivated by the problems in manual attendance systems at schools, which remain vulnerable to fraud, time-consuming, and inefficient. The expected solution is to develop an automated attendance system based on face recognition that can operate in realtime with high accuracy. The research object is vocational high school students, with the applied method implementing the YOLO v10 algorithm for face detection, followed by the face_recognition library for identification. The instruments used include an Imou CCTV camera as the input device, a mid-range laptop as the hardware platform, and Python with SQLite as the software environment for data processing and attendance storage. The results show that the developed system achieved an average face detection accuracy of 96% under normal lighting and 91% under low lighting, with an average processing speed of 27 FPS. The implementation of an anti-duplication feature also ensured data validity by allowing each student to be recorded only once per day. In conclusion, the use of YOLO v10 in face-based attendance proved to be effective, efficient, and capable of reducing fraud. The implication of this study is that the system can be applied in both Islamic boarding schools and general schools as a modernization of attendance systems, with a recommendation for further development through web-based application and cloud database integration.

This study aims to analyze a comparison of coal stockpile volumes using the DJI Mavic 3 Pro Unmanned Aerial Vehicle (UAV) method versus the truck count method at PT. Mitra Barito. Data collection was conducted through aerial photography using a UAV at altitudes of 60 meters and 70 meters, as well as Ground Control Point (GCP) measurements using GPS. The aerial imagery data was processed using photogrammetry software to generate orthophotos and a Digital Elevation Model (DEM), followed by a geometric accuracy test based on the Geospatial Information Agency Regulation No. 6 of 2018, using the Circular Error 90% (CE90) and Linear Error 90% (LE90) parameters. The research results show that high-quality processing at an altitude of 60 meters yields a CE90 value of 2.1619 meters and an LE90 value of 4.3656 meters, thereby meeting the accuracy standards for RBI maps at a scale of 1:5,000, Class 3 for horizontal accuracy, and a scale of 1:10,000, Class 3 for vertical accuracy. Volume calculations of the stockpile using UAVs yielded a result of 22,750.900 m³, while the truck count method produced a volume of 23,503.300 m³. The volume difference between the two methods was 753.400 m³, with a deviation percentage of 3.2%. Based on the research results, the UAV method is considered capable of providing relatively accurate calculations of coal stockpile volume.

This study aims to develop a financial saving application to improve the saving habits of students, particularly in Islamic boarding schools, through an adaptive challenge approach. The system integrates a mobile iOS application with a backend service and Large Language Model (LLM) processing via Ollama. Transaction data entered by users is processed by the backend to generate contextual and personalized saving challenges, applying Reinforcement Learning concepts in an adaptive and data-driven manner. The research adopts a descriptive quantitative method using surveys and system testing with 50 respondents. Results indicate that the application functions as designed, with no significant bugs detected. User evaluation shows high satisfaction, with an average score of 4.3 out of 5, covering ease of use, interface design, and increased awareness of saving. The combination of gamification, reward systems, and adaptive personalization successfully motivates users to save regularly. This system demonstrates the potential of integrating AI-driven personalization to strengthen financial literacy and healthy financial habits among students in a fun and interactive way.methods, and a summary of the results. The abstract should end with a comment about the significance of the results or conclusions brief.

Gesture  recognition  technology  is  used  to  detect  movements  through  image processing,   enabling  computers  or digital devices to understand and interpret human  body  movements  as  input  or  commands.   This  technology  has  great potential  to bridge communication between the deaf community and individuals without   hearing   impairments,    enhancing  interaction  and  enriching  mutual understanding between the two.  However,  the accuracy ofgesture recognition is often  affected  by variations in the distance between hand landmarks.  Based on this problem,  this research proposes a methodfor stabilizing the measurement of distances between landmark points  in gesture recognition through a polynomial regression  approach.   Specifically,   the  distance  between  hand  landmarks  is calculated and stabilized using polynomial  regression to improve the accuracy of gesture recognition.  This method is implemented using the MediaPipeframework to detect and track hands in real-time,  and the OpenCV library to manage video. The  research  results  show  that  this  approach  can  significantly  improve  the stability  and accuracy  of gesture detection.   The developed system successfully detects gestures for  letters A  through F with a high accuracy  rate,  averaging above 98,3%.  The use ofpolynomial regression helps enhance detection accuracy by reducing noise in the landmark data.