Predicting Future Downtime Using Online Reporting on Track for Monitoring Production Process in Dairy Industry Using Machine Learning Algorithm
Abstrak
Manufacturers are currently embracing the notion of a convergence between the cyber and physical world. Manufacturers worldwide are embracing Industry 4.0 standards to promote modern information technology in their factory production processes, leading to the development of Smart Manufacturing. Emerging information technology concepts such as the Internet of Things (IoT), cloud computing, and artificial intelligence (AI) can be strategically utilized and implemented to support Data-Driven Smart Manufacturing. This research develops a web-based application to predict downtime using the Next.js framework. Additionally, the researcher also uses data extracted from a central database in the form of CSV as the primary dataset for the development. For predicting future downtimes, the researcher uses Support Vector Regression (SVR) and Random Forest algorithm as the main algorithm for testing the Machine Learning method. The researcher will also utilize FastAPI to enable communication between front-end and back-end component. This research will contribute to the development of data-driven smart manufacturing by providing a methodology and best practice on how to improve reporting and decision-making based on available data extracted from the production line.