Claim Missing Document
Check
Articles

A Comparative Study of Convolutional Neural Networks and Vision Transformers for Fruit Classification Malik Jawarneh; Arief Marwanto; Dedy Syamsuar; Maivi Kusnandar
International Journal of Advances in Artificial Intelligence and Machine Learning Vol. 2 No. 2 (2025): International Journal of Advances in Artificial Intelligence and Machine Learni
Publisher : CV Media Inti Teknologi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58723/ijaaiml.v2i2.435

Abstract

Background of study:  Accurate fruit classification is vital for agricultural automation, yet traditional methods are often subjective and inefficient. Convolutional Neural Networks (CNNs) are effective but struggle with global context in fine-grained tasks. Vision Transformers (ViTs), inspired by NLP models, offer global attention mechanisms that may improve classification in complex scenarios.Aims and scope of paper: This study compares the performance of EfficientNet-B0 (a CNN model) and ViT-B/16 (a Transformer model) on a fruit classification task involving five fruit types. The goal is to evaluate their strengths and weaknesses under controlled experimental conditions using a moderately sized dataset.Methods: A dataset of 10,000 fruit images was preprocessed with standard augmentation techniques and split into training and validation sets. Both models were fine-tuned using pretrained weights. Performance was evaluated using accuracy, precision, recall, F1-score, and confusion matrices.Result: EfficientNet-B0 achieved higher overall accuracy (94%) than ViT-B/16 (92%). The CNN model performed consistently across all classes, particularly excelling in bananas and strawberries. ViT-B/16 showed superior results for strawberries but struggled with apples. Confusion matrices revealed class-specific strengths and weaknesses.Conclusion: EfficientNet-B0 is better suited for general fruit classification due to its balanced performance, while ViT-B/16 excels in capturing fine-grained visual features. A hybrid approach may leverage both models’ strengths for enhanced performance in real-world applications.
Pemeliharaan Prediktif Berbasis ANFIS untuk Perangkat EKG Portabel Menggunakan Masukan Multi-Sensor Agus Supriyanto; Arief Marwanto
MEDIKA TRADA Vol 7 No 1 (2026): MEDIKA TRADA (JTEMP) Vol 7 No 1 (2026)
Publisher : LPPM POLBITRADA

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59485/yj0ty525

Abstract

Perkembangan peralatan medis portabel seperti Electrocardiograph (ECG) di Rumah Sakit tipe C menghadapi tantangan signifikan dalam hal pemeliharaan preventif akibat keterbatasan data historis dan ketidakmampuan metode konvensional dalam memodelkan hubungan non-linear antara parameter operasional dengan risiko kegagalan. Penelitian ini mengusulkan kerangka kerja predictive maintenance berbasis Adaptive Neuro-Fuzzy Inference System (ANFIS) dengan integrasi multi-sensor yang mencakup suhu, kelembaban, durasi pemakaian, dan umur alat. Data sintetis berjumlah 230 sampel (30 historis + 200 sintetis) dihasilkan menggunakan fungsi keanggotaan sigmoid untuk mengatasi kelangkaan data. Model ANFIS dilatih dengan arsitektur 5 lapisan menggunakan fungsi keanggotaan generalized bell dan algoritma hibrida backpropagation serta least squares. Hasil evaluasi menunjukkan akurasi model mencapai 98,67% pada data uji, dengan presisi 92,3%, recall 98,2%, dan F1-score 95,2%. Perbandingan dengan metode Fuzzy Mamdani (96,67%), Random Forest (91,5%), dan SVM (88,3%) menunjukkan keunggulan signifikan ANFIS dalam menangani kompleksitas data multi-variabel. Nilai AUC sebesar 0,978 mengonfirmasi kemampuan diskriminasi model yang sangat baik. Kerangka kerja ini berkontribusi pada pengembangan sistem prediktif yang adaptif, akurat, dan interpretabel untuk manajemen peralatan medis di fasilitas kesehatan dengan sumber daya terbatas.
ANALISIS KINERJA KENDALI PID, FUZZY MAMDANI, DAN HYBRID FUZZY-PID UNTUK SISTEM KENDALI SUHU LEMARI OBAT FARMASI Aris Sri Widaryanto; Arief Marwanto
MEDIKA TRADA Vol 7 No 1 (2026): MEDIKA TRADA (JTEMP) Vol 7 No 1 (2026)
Publisher : LPPM POLBITRADA

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59485/3a0gd946

Abstract

Penelitian ini membandingkan kinerja kendali suhu PID konvensional, Fuzzy Mamdani, dan Hybrid Fuzzy-PID pada plant termal lemari obat orde dua (G(s)=1/(7200s²+180s+1)) dengan simulasi MATLAB/Simulink pada set point 25°C dan gangguan -5°C. Parameter evaluasi: rise time, settling time, overshoot, deviasi gangguan, recovery time, dan IAE. Hasil: PID dan Hybrid Fuzzy-PID identik (overshoot 5.31%, deviasi 6.33°C, IAE 1072), sedangkan Fuzzy Mamdani standalone inferior (overshoot 53.88%, deviasi 50.20°C, IAE 8836). Fuzzy Mamdani tidak direkomendasikan. Hybrid Fuzzy-PID direkomendasikan untuk implementasi jangka panjang karena adaptif terhadap perubahan parameter plant seiring waktu, meski performa setara dengan PID pada kondisi linear
PENGALIHAN BEBAN OTOMATIS UNTUK TRAFO YANG TIDAK SEIMBANG BERBASIS LOGIKA FUZZY MAMDANY Rangga Bismantara; Arief Marwanto
Jurnal Kajian Teknik Elektro Vol 11, No 1 (2026): JKTE VOL 11 NO 1 (MARET 2026)
Publisher : Universitas 17 Agustus 1945 Jakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52447/jkte.v11i1.8475

Abstract

Load imbalance in a three-phase distribution transformer can lead to increased neutral current and power losses, which negatively affect the efficiency and reliability of the power distribution system. This study aims to determine the level of load imbalance, the amount of power loss due to neutral current, and to evaluate the effectiveness of a two-time-point balancing method (WBP and LWBP) applied to transformer PNIAI007 at PT PLN (Persero) ULP Enarotali. In addition, a simulation using the Mamdani fuzzy logic method was conducted to validate the balancing results based on field measurement data. The results show that the two-time-point balancing method significantly reduced load imbalance from 29% to 2.47% during the day and 1.21% at night. Neutral current decreased from 37.81 A to 14.58 A (daytime) and from 51.76 A to 19.83 A (night). Power losses were reduced by more than 85%. The Mamdani fuzzy simulation supported the accuracy of field data and can serve as a foundation for automatic load evaluation.
Analisis Penyeimbangan Beban Secara Manual Menggunakan Metode Dua Titik Waktu pada Transformator Distribusi 50 kVA PNIAI007 Rangga Bismantara; Arief Marwanto
Jupiter: Publikasi Ilmu Keteknikan Industri, Teknik Elektro dan Informatika Vol. 4 No. 3 (2026): Mei: Jupiter: Publikasi Ilmu Keteknikan Industri, Teknik Elektro dan Informatik
Publisher : Asosiasi Riset Ilmu Teknik Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.61132/jupiter.v4i3.1376

Abstract

Load imbalance in three-phase distribution transformers can increase neutral current and power losses, thereby reducing the efficiency and reliability of electric power distribution systems. This study aims to analyze the level of load imbalance, power losses caused by neutral current, and the effectiveness of manual load balancing using the two-time-point method, namely peak load time (PLT) and off-peak load time (OPLT). The research was conducted on the PNIAI007 distribution transformer at PT PLN (Persero) ULP Enarotali using a quantitative descriptive approach with a case study design. Data were obtained through measurements of phase currents and neutral current, which were then analyzed to compare conditions before and after load balancing.The results show that the load imbalance level was significantly reduced from more than 29% to 2.47% during daytime operation and 1.21% during nighttime operation. The neutral current decreased from 37.81 A to 14.58 A during the daytime and from 51.76 A to 19.83 A at night. In addition, power losses due to neutral current were reduced by more than 85%. These results indicate that the two-time-point load balancing method is effective in improving the efficiency and operational reliability of distribution transformers.
Co-Authors A. M. Harb, Hani Abdul Haris Kuspranoto Abo-Taleb, Ahmed Ade Satria Agung Satrio Nugroho agus sukarno Agus Suprajitno Agus Suprajitno Agus Suprajitno Agus Supriyanto Ahmed Abo-Taleb Ahmed S. Samra Ahsraf Ashraf Khalil Amir, Afandi Anang Putranto Andrianto, Dian Andri Arinan Putra Aris Sri Widaryanto Aser Anou Asri, Mohd Hafizulhadi Mohd ‘Atiq, Muhammad Bakhtiar Indra Kurniawan Berkah Fajar Tamtomo kiono Brama Sakti Handoko Brama Sakti Handoko Danang Hendrawan Daniel Triswahyudi Darojat Yugiantoro Dedi Nugroho Dedy Syamsuar Deshinta Arrova Dewi Diah Arie WK Eka Nuryanto Budi Susila Fajar Husain Asy'ari fajar pujiyanto Fitri Anindyahadi Gunawan Alim Habib ALzaroug Abobaker Wardeko Hadi Pranoto Haikal Satria, Muhammad Hani A. M. Harb Hapsari, Jenny Putri Hariyono, Muhammad Akbar I Nyoman Gede Muliawan Imam Much Ibnu Subroto Imam Much Ibnu Subroto Iska Yanuartanti Iwan Setiawan Jenny Putri Hapsari Kamilah Syed Yusof, Sharifah Kamilah, Sharifah khaled jemah basher Khalifa Mansour Khalifa Kuat Supriyadi Laksamana Rajendra Haidar Azani Fajri Lawrence Adi Supriyono Lukman Abdul Fatah M Satria M. Gaballah, Waleed M. Haider Abu Yazid M. Haikal Satria M. Haikal Satria M. Ulin Nuha Maivi Kusnandar Malik Jawarneh Mochamad Hadi Saputra Mohamad Habib Ahsan Mohammad Alfian Mudzakir Mohd Hafizulhadi Mohd Asri Muhamad Haddin Muhammad Haikal Satria Muhammad Haikal Satria Muhammad Haikal Satria Muhammad Qomaruddin Muhammad Qomaruddin Muhammad Qomaruddin Muhammad Rifai Rifai Muhammad ‘Atiq Munaf Ismail Munaf Ismail Munaf Ismail, Munaf Musab Ali El Nefati Mustaqim Musyahar, Ghoni Nashruddin Anwar Nuha, M. Ulin Praditya, Muhammad Irfan Nur Pramono Mukti Wibowo Putra, Arinan Qirom Qirom Rangga Bismantara Rangga Bismantara Renantivani, Alivia Reza Yoga Diputra Riky Maulana Firdaus Rio Subandi S Sukiran S. Kamilah S. Y S. Kamilah S. Y, S. Kamilah S. Y S. Samra, Ahmed Sarman Sarman Sarman Sarman Sarman, Sarman Satria, M Satria, M. Haikal Satria, Muhammad Haikal Setyanto, Barry Nur Sharifah Kamilah Syed Yusof Sharifah Kamilah Syed Yusof Sharifah Kamilah Syed Yusof Sharifah Yusof Sigit Prakosa Adhi Nugraha Sonny Hady Winoto Sri Arttini Dwi Prasetyawati Sucipto, Dany Sukarno Budi Utomo Sunaryo Sunaryo Sunaryo, Sunaryo Sunu Arsy Pratomo Suprawikno . Supriyadi, Kuat Suryani Alifah Sutarta Tarta Suyanta Suyanta Tri Basuki Kurniawan Triswahyudi, Daniel Vugar Abdullayev Waleed M. Gaballah Yahya Hidayatullah Yazid, M. Haider Abu Yudha Adi Putra Yudistira Marsya Puvindra Yusof, Sharifah Zakariya Ali Saeid Saeid