Data-driven decision support methodology for enhancing production machine availability

Unscheduled preventive maintenance negatively impacts product quality and increases production time due to downtime and emergency shutdowns, raising production costs. We propose a decision support methodology to enhance equipment availability by analyzing historical time to repair (TTR) data using statistical analysis in Minitab. This study analyzed TTR data from seven machines (Filler, Mixer, Blowmould, Labeller, Variopac, Palletizer, and Conveyor) on a production line for 2022. The analysis included both parametric and non-parametric methods, with results presented graphically to summarize statistics like cumulative repair time probability (CRTPR1) and the hazard rate. Using least squares probability fitting, we found that five machines followed an exponential distribution, while the Palletizer and Mixer exhibited log-normal distributions. All machines had about a 63% probability of completing repairs within the meantime to repair (MTTR), except the Palletizer and Mixer, which showed less than 1% probability.