A Peruvian steel-structure manufacturer was losing USD 189,000 a year to material scrap in its plasma-cutting stage, averaging a 7.18% scrap rate across 11 projects, a 2.2-percentage-point gap against the 5% benchmark set by better-positioned regional competitors. The company's internal incident records attributed an estimated 42.2% of this gap to unplanned machine failures and 58.8% to human error during setup and cutting. Planned maintenance and standardized work were designed and implemented as an integrated, quasi-experimental intervention, with a discrete-event simulation model built and calibrated in Arena used as a design tool to test the logical consistency of the redesigned workflow before field deployment. Over a three-month pilot, the scrap rate fell from 7.18% to 5.50% (-23.4%), cutting errors dropped 46.7%, setup errors 40%, and machine failures 22.2% relative to a comparable pre-intervention baseline, while weekly productivity rose 2.37%. A one-tailed Wilcoxon signed-rank test on the six monitored indicators confirmed the improvement was consistent in direction across all of them (p = 0.016). These single-case results suggest that planned maintenance and standardized work, informed by simulation-based process design, may be transferable to the partially manual, resource-constrained metalworking operations typical of developing economies, extending an approach most prior studies have only reported in highly automated settings, pending validation in other companies.

Keywords

Planned maintenance, standardized work, discrete-event simulation, productivity and metalworking industry.