Claim Missing Document
Check
Articles

Found 2 Documents
Search
Journal : TIERS Information Technology Journal

Comparative Analysis of YOLOv5n and YOLOv8n Deep Learning Models for Precision Detection of Klowong Defects in Batik Fabric Hamidi, Rifqi Restu; Herliansyah, Muhammad Kusumawan; Atmaja, Denny Sukma Eka; Sudiarso, Andi
TIERS Information Technology Journal Vol. 6 No. 1 (2025)
Publisher : Universitas Pendidikan Nasional

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.38043/tiers.v6i1.6499

Abstract

This study presents a comparative analysis of two deep learning object detection models, YOLOv5n and YOLOv8n, for the precies identification of Klowong defects in batik fabric. The evaluation was carried out using a custom dataset consisting of 3,138 annotated images, with 921 allocated for testing and containing 1,295 defect instances across nine defect classes. The main findings show that YOLOv8n outperforms YOLOv5n in both speed and accuracy. YOLOv8n achieved a higher F1-score of 0.87 at a lower confidence threshold (0.297), compared to YOLOv5n’s F1-score of 0.86 at a higher threshold (0.46). In addition, YOLOv8n reduced training time significantly (0.320 hours vs. 0.868 hours) and delivered much faster inference speed (2.9 ms/image), nearly three times quicker than YOLOv5n. Although both models performed well in detecting common defects, YOLOv8n showed more stable results on complex defect types. These improvements make YOLOv8n more suitable for real-time applications in batik production environments. Its efficiency and accuracy support the development of fast and reliable automated quality control systems in traditional textile industries. This research emphasizes the importance of using modern lightweight architectures like YOLOv8n to enhance defect detection performance in practical manufacturing settings.
Operating Room Scheduling Optimization Under Surgeon and Nurse Constraints Using Genetic Algorithm Swilugar, Ayu; Herliansyah, Muhammad Kusumawan
TIERS Information Technology Journal Vol. 6 No. 2 (2025)
Publisher : Universitas Pendidikan Nasional

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.38043/tiers.v6i2.7164

Abstract

Operating room scheduling is a complex problem due to the limited availability of surgeons, nurses, and operating rooms, as well as the variability in surgery durations. Inaccurate predictions or scheduling may cause conflicts such as overlapping surgeon schedules, violations of contamination level restrictions, and unavailability of nurses or rooms, ultimately reducing the quality of hospital services. This study integrates multiprocedure surgery duration prediction using machine learning with scheduling optimization based on genetic algorithms. The prediction model considers the American Society of Anesthesiologists (ASA) physical status classification, patient profiles, and sets of surgical procedures variables. Scheduling optimization employs a lexicographic approach with three main objectives: minimizing patient waiting time, nurse overtime, and operating room idle time, while ensuring surgeon presence during critical phases and nurse availability according to shifts. The results show that the Catboost algorithm achieves the best prediction performance. Incorporating the ASA variable reduces prediction errors by 33.880 minutes in MAE and 55.575 minutes in RMSE compared to model without the ASA feature. The optimization model successfully eliminates all scheduling conflicts, ensuring full compliance with medical procedure constraints. Recovery bed utilization remains efficient, with a maximum of five units used, representing less than 50% of the total capacity.