REWS-GAMA (Radiation Early Warning System) : Modul EWS dengan Sistem Evakuasi Cerdas berbasis Algoritma Dijkstra-Modifikasi untuk Mendukung Sistem Proteksi Fisik

  • Jalalludin Mukhtafi
  • Moehammad Alridz Al Farabi Pasha
  • Nazrul Effendy
Keywords: Artificial intelligence, dijkstra algorithm, early warning system, IoT, radiation

Abstract

In the past few years, nuclear security advancements have been aggressively pursued. In addition to Indonesia's efforts with the establishment of the Hypothetical Nuclear Safety Facility (HNPRF), nuclear global organizations such as the IAEA and ICRP have been actively evaluating evacuation safety using current technology. One of these technological implementations is the REWS module, which has been adopted by many countries, including Lebanon. However, conventional REWS tends to rely on manual procedures, requiring experts to conduct separate analysis and validation, which cannot be performed in real-time. This study aims to develop an AIoT-integrated REWS module to enhance evacuation route response and predictive radiation systems. This research consists of two assessments: a hypothetical study (conducted at the HNPRF) and a pilot study (conducted on the 2nd floor at DTNTF). The proposed solution architecture includes two conditions—“normal” and “hazardous”—by inserting a reference location or initial worker position along with the final evacuation destination. This system provides an evacuation route using a modified Dijkstra algorithm (without weight). Finally, the data visualization is transmitted to Pekerja Proteksi Radiasi (PPR) and local workers, allowing them to follow the recommended evacuation path. The pilot simulation results indicate the most effective exit route, leading to the nearest evacuation point, with an accumulated dose of 1.355 × 10⁻³ μSv, a travel distance of 19.49 meters, and an evacuation time of 8.77 seconds when running. The hypothetical simulation at HNPRF involved two start-point scenarios. In the first scenario (start point-1: front door), the optimal route to end point-1 had an accumulated dose of 4.853 × 10⁻³ μSv, a travel distance of 66.38 meters, and an evacuation time of 29.8 seconds while running. Meanwhile, in the second scenario (start point-2: emergency door), the best route led to end-point-3 (emergency gate), with an accumulated dose of 1.236 × 10⁻² μSv, a travel distance of 161.21 meters, and an evacuation time of 72.55 seconds while running.

Published
2025-11-22
Section
Articles