Jurnal Politeknik Negeri Batam (PoliBatam)
Not a member yet
3001 research outputs found
Sort by
Factors Affecting the Profitability of Islamic Commercial Banks in Indonesia for the Period 2021–2023: FACTORS AFFECTING THE PROFITABILITY OF ISLAMIC COMMERCIAL BANKS IN INDONESIA FOR THE PERIOD 2021–2023
No full textThis study is driven by the importance of understanding sharia banking profitability factors. Capital Adequacy Ratio (CAR), Operational Expansion to Operating Income (BOPO), and Non-Performing Financing (NPF) effect return on assets (ROA) for Indonesian Sharia Commercial Banks from 2021 to 2023. Quantitative multiple linear regression study using secondary data from 12 OJK-listed Sharia Commercial Banks yields 36 observations. CAR positively and considerably affects ROA, highlighting the importance of capital adequacy. BOPO has a significant negative impact, proving that operational efficiency boosts profits. NPF has a positive impact on problem financing risk management. The three variables explain 69.2% of ROA variation. The study\u27s theoretical contribution is to enrich the research on the factors influencing Sharia banks\u27 profitability following the pandemic. Bank managers and regulators can use the findings to improve capital structure, cost efficiency, and risk management to boost financial performance.This study is motivated by the importance of understanding the factors that influence profitability in the islamic banking industry. The objective of the research is to analyze the impact of the Capital Adequacy Ratio (CAR), Operational Expenses to Operating Income (BOPO), and Non-Performing Financing (NPF) on Return on Assets (ROA) at Islamic Commercial Banks in Indonesia during the period 2021–2023. A quantitative approach is employed using multiple linear regression analysis based on secondary data from 12 Islamic Commercial Banks listed with the OJK, resulting in 36 observations. The results indicate that CAR has a positive and significant effect on ROA, highlighting the importance of capital adequacy. BOPO has a negative and significant effect, signaling that operational efficiency plays a key role in improving profitability. NPF has a positive and significant effect, which may reflect the effectiveness of managing problematic financing risk. Together, these three variables explain 69.2% of the variation in ROA. The theoretical contribution of this study lies in expanding the literature on the determinants of profitability in Islamic banks post-pandemic. Practically, the findings provide strategic insights for bank management and regulators to strengthen capital structure, improve cost efficiency, and manage risk to enhance financial performance
Minimasi Lead Time Proses Produksi Pada Part Otomotif Tipe BZ 460 RH Menggunakan Value Stream Mapping (VSM) di PT XYZ
No full textPT XYZ merupakan perusahaan manufaktur yang bergerak dibidang industri otomotif. Pada proses produksi di PT XYZ menghasilkan komponen otomotif salah satu produknya adalah part type BZ460 RH. Berdasarkan pengamatan, ditemukan permasalahan pada pemborosan overprocessing, dan transportasi. Pemborosan yang ditemukan berdampak dengan ketidakcapaian target produksi sebesar 279 unit/hari dengan hanya memproduksi 265 unit/hari. Penelitian ini ditujukan untuk mengidentifikasi dan meminimasi lead time yang terdapat aktivitas pemborosan pada proses produksi part BZ 460 RH. Penelitian ini menggunakan metode Value Stream Mapping (VSM) dengan bantuan Value Stream Analysis Tools (VALSAT) yaitu Process Activity Mapping (PAM) untuk menggambarkan aliran nilai dengan melakukan pemetaan aktivitas proses secara detail. Perencanaan perbaikan dilakukan dengan menggunakan analisis 5W+1H. Hasil production lead time sebelum perbaikan pada current state value stream mapping sebesar 139,44 detik/unit dengan nilai Process Cycle Efficiency (PCE) 33,89%. Perbaikan dilakukan dengan mengusulkan modifikasi meja packing dengan rak plastik packing dan melakukan penambahan trolley sesuai kebutuhan. Hasil production lead time setelah perbaikan mengalami penurunan sebesar 26% menjadi 103,75 detik/unit dan terjadinya peningkatan pada nilai PCE menjadi 45,54%. Hasil tersebut membuat target produksi tercapai dengan output produksi sebesar 313 unit/hari. Berdasarkan hasil penelitian dapat disimpulkan bahwa perbaikan yang dilakukan pada proses produksi part BZ 460 RH memiliki pengaruh signifikan terhadap production lead time, PCE, dan output produksi
Green Technology Adoption: A Systematic Review of Key Trends and Challenges
No full textThis study systematically reviews the trends, drivers, and barriers of green technology adoption, synthesizing insights from 81 articles indexed in the Scopus database from 2019 to 2025. Employing the PRISMA framework and bibliometric analysis, the research aims to provide a comprehensive overview of the academic landscape and offer evidence-based guidance for stakeholders. The findings reveal a growing, albeit limited, academic interest, with a research peak in 2024. Geographically, the discourse is led by developed nations and emerging economies, notably China, while research predominantly focuses on high-impact sectors such as transportation, energy, and manufacturing, leaving critical sectors like agriculture under-examined. Furthermore, this review provides a theoretical contribution by mapping empirical findings onto the Green Innovation Cycle and the Stimulus-Organism-Response (S-O-R) model, thereby strengthening the explanatory power of existing frameworks. We identify key challenges spanning infrastructure, policy, and user behavior, and provide specific recommendations for policymakers, industry leaders, and researchers to foster a more equitable and effective green transition. This research serves as a robust scientific foundation for future studies and strategic initiatives to accelerate global green technology adoption
Extreme Learning Machine Method Application to Forecasting Coffee Beverage Sales
No full textSales estimates can be used to set product prices and increase expected profits. Flyover coffee shop Karanganyar does not have a methodical forecasting method to estimate and predict their need/demand for coffee beverage products. Two previous research that used Extreme Learning Machine (ELM) method in other predictions stated that ELM method has high accuracy and fast compilation time. Another research predicted jeans sales using the ARIMA model and produced an accuracy of 17.05% based on the MAPE (Mean Absolute Percentage Error) method. Menstrual cycle prediction using the Long Short-Term Memory (LSTM) method produces a MAPE value of 7.5%. Two advantages of ELM method from two previous research were used as the basis for selecting ELM method used in our study. To help predict sales of coffee beverage menus, this research utilized an artificial neural network method using ELM algorithm. ELM method consists of an input layer and an output layer connected through a hidden layer. Data used for the test was daily sales data for a month. Data used for this study consisted of 215 data samples. Daily sales data at the Flyover coffee shop were collected from June to December 2024. Based on the results and analysis of error values using MAPE method, an average error value was 8.274%. From comparison of original data results and prediction data, an average MAPE error value the best number of features and hidden neurons is 5.65%
Automated Generation of Folklore Short Stories Using T5 Transformer Model
No full textHigh reading interest plays an important role in increasing knowledge and fostering a stronger literacy culture. With the growing access to information and technology, reading interest is also expected to improve through innovative and interactive platforms. However, traditional reading materials often fail to attract younger generations who are more engaged with digital content. To address this challenge, one of the efforts undertaken is the development of a modern platform that provides a collection of short stories enriched with cultural and educational values, tailored to appeal to contemporary readers. This study aims to design and implement a short story generation system using a Transformer-based language model, specifically T5 (Text-to-Text Transfer Transformer). The model is fine-tuned using a curated dataset of folktales from various regions, with the goal of producing relevant, engaging, and coherent narrative texts. The generation process is supported by pre-processing techniques to structure the data into narrative components such as introduction, conflict, climax, and resolution. The generated stories are then evaluated through human evaluation methods, including questionnaires and User Acceptance Testing (UAT), to assess their quality, coherence, engagement, and cultural relevance. This ensures that the system not only produces technically valid texts but also delivers narratives that are meaningful and enjoyable for readers. Ultimately, this study contributes to the promotion of literacy by presenting local wisdom and traditional values from diverse cultures through stories in a more modern, engaging, and accessible format for the younger generation.High reading interest plays an important role in increasing knowledge and fostering a stronger literacy culture. With the growing access to information and technology, reading interest is also expected to improve through innovative and interactive platforms. However, traditional reading materials often fail to attract younger generations who are more engaged with digital content. To address this challenge, one of the efforts undertaken is the development of a modern platform that provides a collection of short stories enriched with cultural and educational values, tailored to appeal to contemporary readers. This study aims to design and implement a short story generation system using a Transformer-based language model, specifically T5 (Text-to-Text Transfer Transformer). The model is fine-tuned using a curated dataset of folktales from various regions, with the goal of producing relevant, engaging, and coherent narrative texts. The generation process is supported by pre-processing techniques to structure the data into narrative components such as introduction, conflict, climax, and resolution. The generated stories are then evaluated through human evaluation methods, including questionnaires and User Acceptance Testing (UAT), to assess their quality, coherence, engagement, and cultural relevance. This ensures that the system not only produces technically valid texts but also delivers narratives that are meaningful and enjoyable for readers. Ultimately, this study contributes to the promotion of literacy by presenting local wisdom and traditional values from diverse cultures through stories in a more modern, engaging, and accessible format for the younger generation
Application of Naïve Bayes Classifiers for Family Risk Identification and Stunting Intervention Planning
No full textStunting remains a significant public health concern influenced by a combination of social, economic, and environmental factors. This study aims to implement the Naïve Bayes algorithm to support the determination of appropriate intervention strategies for families identified as being at risk of stunting in Metro City. Risk data were obtained from the BKKBN Metro City and underwent preprocessing steps, including handling missing values, encoding categorical variables, and feature selection. The dataset was then divided into training, validation, and testing subsets to develop and evaluate models using three Naïve Bayes variants: Gaussian, Multinomial, and Bernoulli. Evaluation metrics of accuracy, precision, recall, and F1-score indicate that the Multinomial Naïve Bayes model achieved the best performance with 99% accuracy, followed by the Bernoulli Naïve Bayes model with 98% accuracy. Both models effectively classified families at risk of stunting with minimal misclassification, while the Gaussian Naïve Bayes variant demonstrated lower performance with an accuracy of 60%. These results highlight the potential of the Naïve Bayes algorithm, particularly the Multinomial and Bernoulli models, as practical and efficient tools to support data-driven decision-making for stunting interventions.Stunting remains a significant public health concern influenced by a combination of social, economic, and environmental factors. This study aims to implement the Naïve Bayes algorithm to support the determination of appropriate intervention strategies for families identified as being at risk of stunting in Metro City. Risk data were obtained from the BKKBN Metro City and underwent preprocessing steps, including handling missing values, encoding categorical variables, and feature selection. The dataset was then divided into training, validation, and testing subsets to develop and evaluate models using three Naïve Bayes variants: Gaussian, Multinomial, and Bernoulli. Evaluation metrics of accuracy, precision, recall, and F1-score indicate that the Multinomial Naïve Bayes model achieved the best performance with 99% accuracy, followed by the Bernoulli Naïve Bayes model with 98% accuracy. Both models effectively classified families at risk of stunting with minimal misclassification, while the Gaussian Naïve Bayes variant demonstrated lower performance with an accuracy of 60%. These results highlight the potential of the Naïve Bayes algorithm, particularly the Multinomial and Bernoulli models, as practical and efficient tools to support data-driven decision-making for stunting interventions
An Intelligent Web-Based Mental Health Management Platform with Rule-Based Music Therapy Recommendation
No full textThis research developed a web-based application for mental health management with an emotional music therapy recommendation feature using Rule-Based Filtering. The system is designed to help individuals recognize and manage emotional conditions caused by life pressures, work stress, and often overlooked psychological issues. A 2023 survey showed that 43% of respondents were concerned about mental health problems, followed by stress at 40%, while 43.8% of parents of teenagers managed their children’s mental health issues independently, 19.2% did not know where to seek help, and 15.4% believed the problems would improve on their own. The system analyzes daily emotional input and weekly PANAS questionnaires to classify moods (Positive, Negative, Mixed, Neutral) based on Positive Affect (PA) and Negative Affect (NA) scores, then recommends relevant music from the database. The technical implementation uses Laravel for the backend and Tailwind CSS for the frontend. Black Box Testing showed 100% functionality. User Acceptance Test (UAT) with 32 respondents resulted in UAT-J 90.25%, UAT-K 90.41%, UAT-R 89.18%, and UAT-A 91.24%. The System Usability Scale (SUS) reached an average score of 85 (very high), while the Net Promoter Score (NPS) was 59.37% (62.50% Promoters), indicating strong user satisfaction and loyalty. This research is expected to help individuals monitor emotional conditions and increase mental health awareness through an innovative music-based approach
Sentiment Analysis of Trending Topics on Social Media X Using Natural Language Processing and LSTM
No full textIn today’s fast-paced digital era, trending news on Social Media X spreads rapidly, influences public opinion, and is often vulnerable to disinformation. This study analyzes netizens’ sentiment towards trending topics on Social Media X using Natural Language Processing (NLP) and a Long Short-Term Memory (LSTM) model. A dataset of 4483 comments was collected across 15 trending topics (Feb–Jun 2025). The preprocessing steps included cleansing, case folding, stopword removal, tokenization, and translation to handle bilingual data. Results show sentiment distribution: 35% positive, 36% negative, and 29% neutral. Model performance varied between 34%–67% accuracy, with precision, recall, and F1-scores indicating that topic sensitivity, language diversity, and data imbalance strongly influenced outcomes. This research contributes to text analytics by providing a baseline model for real-time trending news sentiment analysis in Indonesia, particularly under multilingual and noisy data conditions.In today’s fast-paced digital era, trending news on Social Media X spreads rapidly, influences public opinion, and is often vulnerable to disinformation. This study analyzes netizens’ sentiment towards trending topics on Social Media X using Natural Language Processing (NLP) and a Long Short-Term Memory (LSTM) model. A dataset of 4483 comments was collected across 15 trending topics (Feb–Jun 2025). The preprocessing steps included cleansing, case folding, stopword removal, tokenization, and translation to handle bilingual data. Results show sentiment distribution: 35% positive, 36% negative, and 29% neutral. Model performance varied between 34%–67% accuracy, with precision, recall, and F1-scores indicating that topic sensitivity, language diversity, and data imbalance strongly influenced outcomes. This research contributes to text analytics by providing a baseline model for real-time trending news sentiment analysis in Indonesia, particularly under multilingual and noisy data conditions
Expert System for Early Detection of Postpartum Complications Using Certainty Factor Method
No full textPostpartum complications are one of the main causes of maternal mortality. The objective of this study was to design and build an expert system capable of early detection and providing consultation regarding health complications that occur during the postpartum period using the certainty factor method. The certainty factor approach is utilized to overcome the uncertainty that arises during the diagnosis process by combining the confidence values of each symptom entered by the user. The research methods included needs identification, data collection, knowledge acquisition, knowledge base development, software design, software development, and testing. Needs identification generated 10 data sets, data collection was conducted through literature studies, and knowledge acquisition was obtained through interviews. The knowledge base was compiled based on information from experienced medical personnel. The software design included data flow, interface, and database. The software development resulted in a good early detection expert system. The expert system trial indicated superior performance in identifying health complications during the postpartum period with an accuracy rate of 91%. The ability to recognize positive cases reached 90.9%, and the error rate was 10%, indicating this system is reliable and accurate in decision-making.Postpartum complications are one of the main causes of maternal mortality. The objective of this study was to design and build an expert system capable of early detection and providing consultation regarding health complications that occur during the postpartum period using the certainty factor method. The certainty factor approach is utilized to overcome the uncertainty that arises during the diagnosis process by combining the confidence values of each symptom entered by the user. The research methods included needs identification, data collection, knowledge acquisition, knowledge base development, software design, software development, and testing. Needs identification generated 10 data sets, data collection was conducted through literature studies, and knowledge acquisition was obtained through interviews. The knowledge base was compiled based on information from experienced medical personnel. The software design included data flow, interface, and database. The software development resulted in a good early detection expert system. The expert system trial indicated superior performance in identifying health complications during the postpartum period with an accuracy rate of 91%. The ability to recognize positive cases reached 90.9%, and the error rate was 10%, indicating this system is reliable and accurate in decision-making
Hybrid PSO-XGBoost Model for Accurate Flood Risk Assessment
No full textFlood risk prediction is a crucial step in disaster mitigation. This study optimizes the Extreme Gradient Boosting (XGBoost) algorithm using the Particle Swarm Optimization (PSO) method to improve prediction accuracy. The process includes data cleaning, normalization, and classification of risk levels into low, medium, and high. The XGBoost model is trained both before and after parameter optimization of n_estimators, max_depth, and learning_rate. Before optimization, the model achieved 93% accuracy but struggled to identify minority classes. After optimization with PSO, accuracy increased to 97%, with the recall for the low-risk class improving from 21% to 57%. The optimized model also demonstrated more stable performance compared to Support Vector Machine (SVM) and Random Forest. These findings indicate that the combination of XGBoost and PSO can provide more accurate and efficient flood risk predictions