Fire Early Warning System in a Room Using Fuzzy Logic Tsukamoto Inference Method Based on Internet of Things

Abstract

Fire incidents are among the most common disasters worldwide and frequently result in significant economic losses, property damage, environmental degradation, and loss of human life. In Indonesia, the number of fire incidents increased in 2023, highlighting the need for more effective fire prevention and early detection mechanisms. Many conventional fire detection systems are limited in their ability to provide real-time monitoring and rapid notification, which can delay emergency response efforts and increase the severity of fire-related losses. Therefore, the development of intelligent and automated fire early warning systems has become increasingly important. This research aims to design and implement an indoor fire early warning system by integrating Internet of Things (IoT) technology with fuzzy logic techniques. The proposed system utilizes a flame sensor to detect the presence of fire and an MQ-2 sensor to monitor smoke concentration levels. An ESP32 microcontroller functions as the central processing unit, collecting sensor data, processing information, and transmitting results to users. To improve decision-making accuracy under uncertain environmental conditions, the Tsukamoto fuzzy logic method is employed to determine the level of fire risk based on sensor inputs. The system is equipped with a real-time notification feature that automatically sends warning alerts through the Telegram application whenever potential fire hazards are detected. In addition, sensor readings and fire status information can be monitored remotely through a web-based platform. Experimental results demonstrate that the proposed system effectively detects fire and smoke conditions while providing timely alerts. The integration of IoT and fuzzy logic enables rapid response and supports proactive fire management, thereby reducing potential losses and enhancing indoor safety.