Fahreza, Dimas Ahmad (2026) Perancangan dan Implementasi Sistem Monitoring Kepatuhan APD Otomatis Berbasis Telegram Bot Menggunakan Arsitektur AI Real-Time. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 5025221319_Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5025221319_Undergraduate_Thesis.pdf Restricted to Repository staff only Download (11MB) | Request a copy |
Abstract
Pengawasan kepatuhan Alat Pelindung Diri (APD) di lingkungan industri dan konstruksi masih mengandalkan pengawasan manual yang tidak efisien, lambat dalam pelaporan, dan rentan terhadap kesalahan manusia. Penelitian ini merancang dan mengimplementasikan sistem pemantauan kepatuhan APD otomatis yang mengintegrasikan pipeline deteksi berbasis kecerdasan buatan dengan arsitektur Telegram Bot interaktif sebagai media pelaporan real-time. Sistem menggunakan model deteksi objek YOLOv8 untuk mendeteksi pekerja, helm keselamatan, dan rompi reflektif, serta algoritma pelacakan multi-objek BoT-SORT untuk mempertahankan identitas unik setiap pekerja antar frame. Pelanggaran APD ditentukan melalui logika berbasis aturan menggunakan perhitungan Intersection over Union (IoU), dengan pemfilteran zona pemantauan menggunakan deteksi berbasis poligon melalui library Shapely. Setiap pelanggaran dicatat secara otomatis ke dalam file CSV harian, kemudian diagregasi menggunakan Pandas untuk menghasilkan laporan berformat grafik PNG dan dokumen PDF yang dikirimkan kepada pengawas melalui Telegram Bot. Pengujian sistem menunjukkan bahwa seluruh komponen pipeline berfungsi sesuai spesifikasi. Tingkat keberhasilan pengiriman laporan mencapai 100% pada kondisi jaringan stabil dan sedang, dengan waktu pemrosesan rata-rata 45–90 detik dari penerimaan video hingga laporan terkirim. Dibandingkan dengan pelaporan manual yang membutuhkan 30–60 menit, sistem ini menawarkan peningkatan efisiensi yang signifikan tanpa memerlukan infrastruktur server tambahan maupun biaya lisensi perangkat lunak.
====================================================================================================================================
Manual supervision of Personal Protective Equipment (PPE) compliance in industrial and construction environments remains inefficient, slow in reporting, and prone to human error. This research designs and implements an automated PPE compliance monitoring system that integrates an AI-based detection pipeline with an interactive Telegram Bot architecture for real-time reporting. The system employs YOLOv8 for detecting workers, safety helmets, and reflective vests, alongside the BoT-SORT multi-object tracking algorithm to maintain consistent identity across video frames. PPE violations are determined through a rule-based logic using Intersection over Union (IoU) calculations, with monitoring zone filtering implemented via polygon-based detection using the Shapely library. Each violation is automatically logged into daily CSV files, then aggregated using Pandas to generate PNG chart visualizations and PDF reports delivered to supervisors through the Telegram Bot. System evaluation demonstrates that all pipeline components perform according to specification. Report delivery achieved a 100% success rate under stable and moderate network conditions, with an average processing time of 45–90 seconds from video receipt to report delivery. Compared to manual reporting that requires 30–60 minutes per report, the system offers a significant improvement in efficiency while requiring no additional server infrastructure or software licensing costs.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Pemantauan Kepatuhan APD, YOLOv8, BoT-SORT, Telegram Bot, Pelaporan Otomatis, Deteksi Real-Time, Keselamatan Kerja, PPE Compliance Monitoring, YOLOv8, BoT-SORT, Telegram Bot, Automated Reporting, Real-Time Detection, Workplace Safety. |
| Subjects: | T Technology > T Technology (General) > T55 Industrial Safety |
| Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis |
| Depositing User: | Dimas Ahmad Fahreza |
| Date Deposited: | 21 Jul 2026 05:47 |
| Last Modified: | 21 Jul 2026 05:47 |
| URI: | http://repository.its.ac.id/id/eprint/135857 |
Actions (login required)
 |
View Item |