Fog Computing Concept Implementation in Work Error Detection System of The Industrial Machine Using Support Vector Machine (SVM)

The implementation of industry 4.0 to integrate various infrastructures is inseparable from the application of sensors, which is then followed by intensive data analysis. One form of the applications is the work error detection system for the industrial machine with the Fog Computing concept. In this study, the Fog Computing architecture was studied to support infrastructure integration for the development of smart industries. This study covers sensor network architecture, device communication, and smart computing with a robotic arm infrastructure. The sensor is installed to monitor the work process of the robot arm; then, the measurement values obtained sent to the fog device for analysis. The results of the analysis to determine the movement patterns then classified by the Support Vector Machine (SVM) method. The result of the system is that it can detect the movement of a robot arm. The test results of this study showed an average accuracy of 90%. Therefore this research can later be used as an initial demonstration and real learning to applied to other machine components and services in supporting the application of smart industries, especially in Indonesia, in the future.