Abstract
Emergencies, such as a fire, cause harm to a person and his property. Reducing the number of losses is a priority aspect for countries and citizens. The main purpose of this research is to develop a fire warning and detection system, which includes a hardware part for collecting data from sensors and a software part for monitoring the environment. The hardware based on the Internet of Things (IoT) technology consists of a NodeMCU V3 microcontroller and flame, humidity, temperature, pressure, and gas sensors. Readings from sensors, as well as notification of a fire threat can be received from anywhere in the world thanks to a monitoring system and a microcontroller. A key feature of the decision-making system in case of fire is the ability to notify and send SMS to social networks. If deviations from the specified range are detected, a voice warning will sound, the received message will include geolocation of the accident site. The collected database from sensors is used for machine learning methods to reduce the likelihood of false calls and alerts. With the correct implementation of the decision-making system, it will be possible to protect a person from death.