Unplanned equipment downtime remains one of the most persistent and costly challenges in food manufacturing, yet the metrics used to manage it are not always well understood in relation to broader production outcomes. Mean Time to Repair (MTTR) is a widely used maintenance indicator, but its specific influence on Overall Equipment Effectiveness (OEE), the composite measure of availability, performance, and quality, has received limited empirical attention, particularly in continuous food production settings. This paper examines that relationship using operational data from a food manufacturing facility in the Western Cape, South Africa. The facility runs eight machines on a 24-hour production schedule, and 91 days of maintenance and production records were analysed. Descriptive statistics and box-and-whisker plots were used to characterise breakdown frequency, machine availability, and OEE variability across the study period. Because the MTTR-OEE relationship showed a nonlinear pattern, Locally Weighted Scatterplot Smoothing (LOWESS) regression was applied to model it without imposing a fixed functional form. The findings confirm that repair duration exerts a measurable, nonlinear effect on OEE, with prolonged repair times disproportionately eroding availability and overall equipment performance. These results carry practical implications for maintenance planning: reducing MTTR through faster diagnostics, better parts availability, or operator-level intervention offers a low-capital route to meaningful OEE gains. The study contributes empirical evidence to the lean manufacturing and Total Productive Maintenance literature, and provides a replicable analytical framework for facilities seeking to improve equipment effectiveness through targeted maintenance improvements.

Keywords

Lean manufacturing, Total Productive Maintenance (TPM), Overall Equipment Effectiveness, Meantime to Repair (MTTR) and Availability losses.