Claim Missing Document
Check
Articles

Found 3 Documents
Search

Menghitung Kendaraan Di Parkiran Dengan Metode Support Vector Machine (SVM) Masnur; Syahirun Alam; Muhammad Zainal; asriadi asriadi
Jurnal Ilmiah Teknik Unida Vol. 3 No. 2 (2022): Des
Publisher : Mitra Teknik

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55616/jitu.v3i2.369

Abstract

Support Vector Machine (SVM) merupakan sebuah metode klasifikasi yang digunakan untuk mengklasifikasi data dan bekerja dengan cara mendefinisikan batas antara dua kelas dengan jarak maksimal dari data yang terdekat. Untuk mendapatkan batas maksimal antar kelas maka harus dibentuk sebuah hyperplane (garis pemisah) terbaik pada input space yang diperoleh dengan mengukur margin hyperplane dan mencari titik maksimalnya. Pada paper ini akan dijelaskan mengenai pembuatan aplikasi pendeteksi objek dengan metode klasifikasi yaitu SVM. Menggunakan polyline sebagai cropping untuk dataset dan mengambil kordinat frame yang akan disimpan di file bentuk JSON sebagai pada saat menjalankan aplikasi kotak yang dibuat tersimpan dalam bentuk frame Pada penelitian ini penulis dapat menghasilkan sebuah aplikasi untuk mendeteksi objek khusunya menghitung menghitung kendaraan di parkiran dengan menggunakan metode Support Vector Machine (SVM) dalam proses pengklasifikasian dataset untuk menentukan atau membedakan kendaraan yang terparkir dan yang tidak.
Rancang Bangun Sistem Manajemen Produksi dan Distribusi di Konveksi Berbasis Web App (PWA) Fajri Alamsyah; Ahmad Selao; Andi Wafiah; Masnur; Syahirun Alam; Muh. Zainal
Jurnal Informatika dan Komputer Vol 16 No 1 (2026): April
Publisher : Sekolah Tinggi Ilmu Komputer PGRI Banyuwangi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55794/jikom.v16i1.326

Abstract

This study develops a Progressive Web App (PWA)-based production and distribution management system for garment industries, aiming to enhance operational efficiency and production process transparency. The designed system integrates all production stages—from raw material planning, production scheduling, task allocation, to quality control—featuring real-time monitoring accessible across multiple devices. Using a quantitative approach with development research design, the study involved 30 end-users (10 production managers and 20 garment workers). Data was collected through Likert-scale questionnaires and in-depth interviews, analyzed via descriptive statistics and thematic methods. Evaluation results show average scores of 4.2/5 for usability, 3.8/5 for system security, and 4.0/5 for user satisfaction. Technical testing demonstrated average response time of 1.2 seconds, load time of 2.8 seconds, and offline capability up to 72 hours. Key findings reveal that 85% of respondents rated the app as user-friendly with an intuitive interface, while 78% reported improved team coordination efficiency. However, technical challenges were identified, particularly in data security (only 65% found security features adequate) and limited analytics for strategic decision-making. This research provides practical contributions through affordable technology solutions for small-scale garment businesses, while identifying development areas: (1) multi-factor authentication integration, (2) enhanced offline data backup systems, (3) service worker caching strategy optimization.
Implementasi Sistem Prediksi Pemesanan Tiket Online Untuk Optimalisasi Penjualan Menggunakan Random Forest Bintang Choirul Nusri; Mughaffir Yunus; Masnur; Nurdiansyah Sirimorok; Syahirun Alam; Muh. Zainal
Jurnal Informatika dan Komputer Vol 16 No 1 (2026): April
Publisher : Sekolah Tinggi Ilmu Komputer PGRI Banyuwangi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55794/jikom.v16i1.327

Abstract

The online ticketing industry faces challenges in managing fluctuating ticket demand. This study aims to develop an online ticket booking prediction system using the Random Forest algorithm to optimize ticket sales. Using historical ticket booking data, a predictive model is built to project future ticket demand based on variables such as ticket price, booking time, event type, and event location. The data used includes 5,000 randomly selected ticket transactions from online ticketing service providers. The results show that the Random Forest model provides more accurate predictions compared to baseline methods (linear regression and single decision tree). The model achieved MAE of 0.142, RMSE of 0.185, and R² of 0.892, showing significant improvement compared to linear regression (MAE: 0.321; RMSE: 0.398; R²: 0.642) and single decision tree (MAE: 0.218; RMSE: 0.285; R²: 0.754). Statistical testing using paired t-test showed significant difference (p-value < 0.001) between Random Forest and baseline models. These findings indicate that a Random Forest-based prediction system can help ticket providers optimize pricing, inventory management, and ticket sales efficiency, and open up opportunities for the model's application in other sectors.