Predictive Maintenance of Relative Humidity Using Random Forest Method
Aji Teguh Prihatno, Himawan Nurcahyanto, Yeong Min Jang
Abstract
The massive development of Industry 4.0 inseparable with improvement of Machine Learning. In order to protect manufacturing sector from unwanted events such as electrical failures due to high level of humidity, the predictive maintenance based on Machine Learning should be developed accurately. This paper describes the implementation work of predicting Relative Humidity (RH) in the smart factory's environment by using Random Forest method as a part of Machine Learning. In order to support data reliability and interoperability in smart factory environment, IIoT devices based oneM2M standard platform was used to collect the data. The result of this Random Forest method for predict relative humidity shows 82.49% which considered as an excellent accuracy. This research goal may contribute to the manufacturing fields to be able to lower the cost and increase efficiency in maintenance.