Jurnal Politeknik Negeri Batam (PoliBatam)
Not a member yet
3001 research outputs found
Sort by
Evaluation of a Virtual Reality-Based Introduction to Hazardous and Toxic Waste Management Using the Technology Acceptance Model
Get PDFThe management of hazardous and toxic waste (B3 waste) demands innovative educational approaches to improve technical comprehension and environmental awareness, particularly among internship students. This study presents a formative evaluation of an early-stage virtual reality (VR) application prototype, aimed at assessing initial user perceptions and gathering feedback to guide further development, using the Technology Acceptance Model (TAM) framework. The prototype was tested on six internship students involved in the supervision of B3 waste in collaboration with the Environmental Agency (Dinas Lingkungan Hidup) of East Java Province. Data were collected through a questionnaire focusing on three TAM dimensions: Perceived Ease of Use, Perceived Usefulness, and Attitude Toward Technology. The results showed that the VR application was perceived as highly useful (87.5%), easy to use (89.2%), and positively received by users (92.5%). These findings indicate that VR technology holds strong potential as an interactive learning tool for introducing hazardous and toxic waste management practices. The study recommends continued content development and broader testing with a larger respondent base to validate these initial results
Real-Time Braille Letter Detection System Using YOLOv8
Get PDFThe purpose of this research is to create a system capable of detecting and recognizing Braille letters in real-time using the YOLOv8 algorithm for object detection, integrated with image processing technology and a user interface based on Tkinter. This system is developed to support visually impaired individuals in reading Braille text through the use of a webcam that captures and identifies Braille letters from images. The identification process is carried out by comparing the obtained images with a precompiled database of Braille letters. This research utilizes a dataset consisting of images of Braille code from letters A to Z, collected through public and private methods, with a total of 6013 images that comprehensively represent Braille letters. The model training is done using YOLOv8 to recognize Braille letter objects, with model performance evaluation using the Mean Average Precision (mAP) metric.The results of the tests show a very satisfactory model performance, with a mAP50 score of 0.98 and a mAP50-95 score of 0.789, as well as a high accuracy rate for almost all Braille letters tested. In addition, the system is equipped with a Tkinter-based graphical user interface (GUI) that allows users to operate the Braille letter detection process interactively and easily. This research proves that the YOLOv8-based object detection approach has significant potential for Braille letter recognition applications, which is expected to enhance accessibility and the independence of visually impaired individuals in reading text effectively
Implementation of Conditional WGAN-GP, ResNet50V2, and HDBSCAN for Generating and Recommending Traditional Lombok Songket Motifs
Get PDFSongket is a traditional Indonesian woven textile with profound cultural and aesthetic value, particularly in Lombok, where artisans continue to preserve its distinctive motifs. However, the creation of new designs is still carried out manually, requiring considerable time and relying heavily on the artisans’ creativity. This study proposes an integrated system that combines Conditional Wasserstein Generative Adversarial Network with Gradient Penalty (CWGAN-GP), ResNet50V2, and HDBSCAN to automatically generate and recommend Lombok’s traditional songket motifs. The dataset consists of primary data collected directly from local artisans and secondary data from the BatikNitik public repository, thereby providing authentic yet diverse motif samples for training. CWGAN-GP is employed to synthesize motifs with stable and realistic structures across multiple epochs. Subsequently, ResNet50V2 is utilized for deep visual feature extraction, HDBSCAN for density-based clustering, and UMAP for two-dimensional visualization of motif distribution. The system successfully groups motifs into meaningful clusters, with the largest cluster containing consistent patterns of high aesthetic value. A recommendation mechanism is also developed to suggest up to five similar motifs from the original dataset within the same cluster, ensuring cultural relevance while fostering design innovation. Despite these promising outcomes, several limitations remain, such as the relatively small number of songket motif samples, variations in motif quality, and challenges during data collection including inconsistent lighting and non-uniform patterns. These factors affect both dataset consistency and generative performance. Nevertheless, this approach demonstrates the potential of artificial intelligence to support the preservation and innovation of cultural heritage by assisting artisans in creating and exploring new motifs more efficiently without losing their traditional identity
Proposed Improvement of the Supplier Selection Process using Analytical Hierarchy Process (AHP) Method: A Case Study of Aircraft MRO in Indonesia
Get PDFIndonesia’s aviation industry has experienced rapid post-pandemic growth, with a 38.5% increase in passenger traffic in 2024 compared to 2022, positioning it as the largest and fastest-growing market in ASEAN. This expansion has intensified demand for Maintenance, Repair, and Overhaul (MRO) services. PT XYZ, Indonesia’s leading aircraft maintenance provider, faces procurement issues in acquiring painting materials, causing delays and missed turnaround targets. This study aims to optimize supplier selection through the Analytical Hierarchy Process (AHP), employing a mixed-methods approach involving stakeholder interviews to determine key criteria. The model incorporates six main criteria and fourteen sub-criteria, prioritizing quality, regulatory compliance, reliability, cost, and delivery. The results identified Supplier II as the most suitable option. The study recommends institutionalizing the AHP model within PT XYZ’s procurement policy and integrating it into its ERP system, offering a scalable solution for enhancing procurement efficiency and supporting strategic decision-making in aviation MRO operations
RANCANG BANGUN BRACKET PLATFORM PADA MESIN WELD MANIPULATOR
No full textA welding manipulator is a specialist apparatus engineered for welding extensive and elongated pipe plates on both interior and exterior surfaces, as well as in vertical and horizontal positions. During the assembly process, a challenge occurs when the platform is excessively distanced from the nozzle, complicating the operator\u27s task with the welding manipulator. Brackets are an effective option to mitigate hazards and address issues throughout the cutting process. The aim of this research is to design and evaluate the safety of the bracket utilized for the platform component of the welding manipulator. The research utilizes Finite Element Analysis (FEA) to assess the simulated values of the safety factor and stress distribution. The findings demonstrate that the bracket design satisfies strength criteria, producing a maximum stress value of 98,327,504.00 N/m². The recorded maximum displacement is 1.094 mm, and the maximum strain is 0.00022. The computed safety factor is 2.1, indicating that the design parameters surpass established safety criteria.Welding manipulator adalah jenis alat pengelasan yang digunakan untuk pengelasan pada plat pipa berukuran besar dan panjang pada bagian dalam dan luar dalam bentuk vertikal dan horizontal. Terdapat masalah ketika peletakan platform jauh dari nozzle selama proses assembling yang menyebabkan masalah bagi operator weld manipulator. Oleh karena itu bracket menawarkan solusi untuk masalah dan risiko yang muncul selama proses pemotongan sehingga memungkinkan proses pemotongan dilakukan dengan lebih aman dan efisien. Tujuan penelitian adalah untuk mendesain dan mengetahui hasil analisis keamanan bracket alat pengelasan weld manipulator bagian platform. Untuk menentukan nilai simulasi faktor keamanan dan distribusi tegangan, penelitian menggunakan Finite Element Analysis (FEA). Hasil penelitian menunjukkan bahwa desain bracket memenuhi syarat dari segi kekuatan dengan nilai stress maksimum 98.327.504,00 N/m2; nilai maksimum displacement 1,094 mm dan nilai maksimum strain 0,00022, dan faktor keamanan 2,1 sehingga dapat disimpulkan kondisi berada di atas standar
Desain Alat Pengering Pozzolan di PT Semen Padang
No full textPT. Semen Padang, sebagai salah satu perusahaan semen terus berinovasi untuk meningkatkan kualitas produk dalam proses produksinya. Salah satu inovasi yang dikembangkan yaitu alat pengering pozzolan akan digunakan untuk mengeringkan material pozzolan yang mengandung kadar air 18%. Tujuan utama untuk menghasilkan sistem pengering yang efisien, andal, dan sesuai dengan karakteristik material pozzolan yang digunakan di PT Semen Padang. Alat pengering pozzolan didesain menggunakan metode flash drayer dengan bantuan software solidworks yang mampu menyediakan sketsa 2D dan dapat di-upgrade menjadi bentuk 3D, serta juga bisa digambarkan dalam sebuah drawing. Spesifikasi alat pengeringan pozzolan yang dihasilkan panjang alat 2400 mm dan lebar casing 315 mm, massa material pozzolan 60 ton/jam kecepatan rantai yang dibutuhkan 0,03 m/sec dan massa pozzolan pada masing masing rantai scraper 1.333 kg. Alat ini mampu menghilangkan massa air sampai 78 ton sehingga dapat mencapai 5% kadar air. Desain alat pengering pozzolan yang dirancang dapat mengatasi terjadi kerusakan dan penyumbatan di bagian penyaringan siklon atau filter nantinya
Perbandingan metode PID dan Fuzzy Logic Control dalam sinkronisasi dua motor DC
No full textSinkronisasi kecepatan dua motor DC merupakan aspek penting dalam berbagai aplikasi industri, seperti sistemkonveyor, robotika, dan otomasi, karena keseragaman gerak sangat dibutuhkan. Penelitian ini membandingkan dua metode pengendali, yaitu Proportional-Integral-Derivative (PID) dan Fuzzy Logic Controller (FLC), dalam menjaga sinkronisasi kecepatan antara dua motor DC. Sistem dirancang agar motor kedua dapat mengikuti kecepatan motor pertama dengan akurasi tinggi, dan masing-masing metode diuji pada skenario perubahan kecepatan yang sama. Evaluasi dilakukan berdasarkan parameter kinerja, seperti waktu tunda (delay time), waktu tunak (settling time), serta kesalahan sinkronisasi terhadap kecepatan target dengan toleransi ±2%. Hasil pengujian menunjukkan bahwa metode FLC memiliki keunggulan dalam kecepatan respons dan kemampuan adaptasi terhadap perubahan dinamis, sehingga menghasilkan waktu sinkronisasi yang lebih cepat serta kesalahan yang lebih kecil dibandingkan dengan metode PID. Sebaliknya, metode PID memberikan kinerja yang stabil pada kondisi beban tetap, tetapi kurang tanggap terhadap gangguan dan perubahan mendadak. Temuan ini membuktikan bahwa logika fuzzy lebih efektif digunakan pada sistem yang memerlukan respons cepat dan fleksibel terhadap kondisi variatif. Dengan demikian, metode FLC direkomendasikan untuk aplikasi pengendalian motor DC multikanal yang menuntut ketelitian sinkronisasi tinggi serta kemampuan adaptif terhadap ketidakpastian sistem
Analisis dan Perancangan Antarmuka Pengguna Situs Web PT. ACW Tour dan Travel Menggunakan Metode Design Thinking
No full textPT. ACW Tour and Travel is a company engaged in the field of tourism services that still uses traditional promotion methods. This study aims to design the UI/UX of the PT. ACW Tour and Travel website using the Design Thinking approach, in order to provide a better user experience and support digital marketing strategies. The design process includes five stages, namely empathize, define, ideate, prototype, and test. Data were collected through interviews and questionnaires to potential users. The final result is a website prototype design that was tested using the System Usability Scale (SUS) method with an average score of 77.3, which is categorized as "very good" and "acceptable". These results indicate that the Design Thinking approach is effective in creating a UI/UX that is in accordance with user needs and expectations.PT. ACW Tour and Travel merupakan perusahaan yang bergerak di bidang jasa pariwisata yang masih menggunakan metode promosi tradisional. Penelitian ini bertujuan untuk merancang UI/UX website PT. ACW Tour and Travel dengan menggunakan pendekatan Design Thinking, agar dapat memberikan pengalaman pengguna yang lebih baik dan mendukung strategi pemasaran digital. Proses perancangan meliputi lima tahap, yaitu empathize, define, ideate, prototype, dan test. Pengumpulan data dilakukan melalui wawancara dan kuesioner kepada calon pengguna. Hasil akhir berupa desain prototipe website yang diuji menggunakan metode System Usability Scale (SUS) dengan skor rata-rata 77,3 yang masuk dalam kategori "sangat baik" dan "cukup baik". Hasil ini menunjukkan bahwa pendekatan Design Thinking efektif dalam menciptakan UI/UX yang sesuai dengan kebutuhan dan harapan pengguna
Public Sentiment Analysis of the Free Nutritious Meals Program (MBG) on Social Media X Using the Naive Bayes Method
No full textThis study aims to analyze public sentiment towards the Free Nutritious Meals Program (MBG) launched by the government, utilizing data from the X (Twitter) platform using the Naïve Bayes method. The background of this study is based on the high level of public attention towards the MBG program, which targets school children, toddlers, pregnant women, and nursing mothers, as well as the prevalence of diverse opinions on social media. Data was collected through a crawling process during the period of April 28 to May 28, 2025, using keywords related to MBG, resulting in 12,310 tweets. The data processing stages included text preprocessing (cleansing, case folding, tokenizing, filtering, stemming), word weighting with TF-IDF, training and test data division, and testing using a confusion matrix. The results show that the Naïve Bayes method is capable of classifying sentiment into three categories: positive, negative, and neutral, with optimal performance on an 80:20 data split, resulting in an accuracy of 86.78%, precision of 86.86%, recall of 86.78%, and an F1-score of 86.58%. The majority of public sentiment towards the MBG program was positive, reflecting support for the program\u27s benefits in improving the nutrition of school children and alleviating the economic burden on families. This study is expected to serve as a reference for the government in evaluating public policy and communication strategies, as well as contributing academically to the development of text mining and sentiment analysis studies on social media.This study aims to analyze public sentiment towards the Free Nutritious Meals Program (MBG) launched by the government, utilizing data from the X (Twitter) platform using the Naïve Bayes method. The background of this study is based on the high level of public attention towards the MBG program, which targets school children, toddlers, pregnant women, and nursing mothers, as well as the prevalence of diverse opinions on social media. Data was collected through a crawling process during the period of April 28 to May 28, 2025, using keywords related to MBG, resulting in 12,310 tweets. The data processing stages included text preprocessing (cleansing, case folding, tokenizing, filtering, stemming), word weighting with TF-IDF, training and test data division, and testing using a confusion matrix. The results show that the Naïve Bayes method is capable of classifying sentiment into three categories: positive, negative, and neutral, with optimal performance on an 80:20 data split, resulting in an accuracy of 86.78%, precision of 86.86%, recall of 86.78%, and an F1-score of 86.58%. The majority of public sentiment towards the MBG program was positive, reflecting support for the program\u27s benefits in improving the nutrition of school children and alleviating the economic burden on families. This study is expected to serve as a reference for the government in evaluating public policy and communication strategies, as well as contributing academically to the development of text mining and sentiment analysis studies on social media
Orchid Species Classification Using the DenseNet121 Deep Learning Model with a Data Imbalance Handling Approach
No full textFor conservation, commercial cultivation, and scientific research, accurate identification of orchid species often requires specialized expertise. In this study, the DenseNet121 deep learning architecture was employed to develop an automated classification system for four popular orchid species. DenseNet121 was selected for its ability to extract complex hierarchical features and its strong performance on limited-scale datasets. The initial dataset comprised 1,935 images of Phalaenopsis, Cattleya, Dendrobium, and Vanda orchids. However, after manual removal of duplicate images, only 1,658 images remained, revealing significant class imbalance. The undersampling method was applied to balance each class to 248 samples. The dataset was then split into 75% training, 15% validation, and 10% testing, and enhanced through data augmentation techniques such as rotation, flipping, brightness variation, width shift, height shift, and zoom. The final model achieved 97.00% accuracy with class-specific performance ranging from 92.59% to 100% accuracy across different orchid species. This research can serve as a foundation for developing mobile or web applications to assist researchers, farmers, and orchid enthusiasts in accurately identifying orchid species, while supporting conservation efforts for orchid biodiversity in Indonesia