The Safety Management System Using Q-Learning Algorithm in IoT Environment

The Industrial Internet of Things (IIoT) is the framework in which a large number of devices are connected and synchronized for handling different processes and machinery in industry, to remove the risk of human error and improve safety. Many IoT-based worker's safety systems contain a network of different sensors and assessable mobile information center. Such as, workers wear sensors that monitor the heart rate, activity, toxic gases, and other factors that are affecting worker's safety. The initial investment for such automation is very high and it is not affordable for small industries therefore, mostly high-level manufacturing industries install automation in the workplace. The purpose of this paper is to design a safety management architecture for workers in the small-scale candy manufacturing industry using IoT and Q-learning mechanism. Also, to provide the benefits of automation at a low-cost. The system consists of gas, temperature, humidity, and flame sensor. ADC is used to convert recorded sensor values from analog to digital form and these converted values are received by the raspberry pi board and simultaneously stored in a database. The Q-learning approach is used to identify crucial situations using database values and execute the output appliances like a fan and buzzer.