<![CDATA[Abstract. Fires frequently occur due to human negligence or electrical system failures, potentially causing significant losses, particularly in high-activity environments such as homes, offices, and archive rooms (Nento et al., 2021). Conventional alarm systems are ineffective in providing remote warnings, as they rely solely on sirens or lights, which become useless when the location is unoccupied (Maulana, 2025). This study developed an Internet of Things (IoT)-based fire detection system capable of delivering early warnings in real-time via a Telegram Bot. The system was designed using an ESP32 microcontroller integrated with three sensors: the MQ-7 for detecting carbon monoxide (CO), the DHT22 for monitoring temperature and humidity, and the IR Flame Sensor for detecting flames. Data from the three sensors is processed by the ESP32 and transmitted via HTTPS using the UniversalTelegramBot library to Telegram, displayed on a 20x4 LCD, and recorded in a Google Spreadsheet (Risanty & Sopiyan, 2017). Testing was conducted at the Department of Tourism and Culture of Bogor City in a simulation room measuring 1.5 m × 1 m × 2 m. The results showed an average system response time of under 5 seconds, with a detection accuracy of 96% and a false positive rate of 2.4%. The system is power-efficient, cost-effective, and easy to implement in high-risk environments such as archive rooms or offices, enabling fast and accurate remote monitoring. This technology offers a significant preventive solution in fire mitigation efforts (Iskandar Mulyana et al., 2023).]]>
Copyrights © 2025
