LAUSR

Abstract Zero Defect Manufacturing (ZDM) is becoming one of the ways that EU can answer to the challenges of the digitization and globalization of industrial manufacturing. There is a growing demand for high quality products at optimal cost. In this work, we consider the manufacturing environment at DANOBAT, one of the largest machine tool and production system manufacturers in Europe. Especially, we focus on the prediction of the Remaining Useful Life (RUL) of grinding machines, which are critical to the manufacturer, as they are responsible for the production and handling of high value material. The RUL analysis can provide valuable information for the deterioration rate of assets; defined as the length from the current time to the end of the useful life. This paper presents the methodology followed to apply the RUL analysis on a real industrial environment, utilizing the Long Short-Term Memory (LSTM) Neural Network algorithm.