Measuring the production performance indicators for metal-mechanic industry: an LDA modeling approach

Abstract

PurposeThis study proposes a novel method to improve the accuracy of overall equipment effectiveness (OEE) estimation in the metallurgical industry. This is achieved by modeling the frequency and severity of stoppage events as random variables.Design/methodology/approachAn analysis of 80,000 datasets from a metal-mechanical firm (2020–2022) was performed using the loss distribution approach (LDA) and Monte Carlo simulation (MCS). The data were further adjusted with a product price index to account for inflation.FindingsThe variance analysis revealed supporting colleagues (59.8% of variance contribution), food breaks (29.8%) and refreshments (9.0%) as the events with the strongest influence on operating losses.Research limitations/implicationsThis study provides a more rigorous approach to operational risk management and OEE measurement in the metal-mechanical sector. The developed algorithm supports the establishment of risk management guidelines and facilitates targeted OEE improvement efforts.Originality/valueThis research introduces a novel OEE estimation method specifically for the metallurgical industry, utilizing LDA and MCS to improve accuracy compared to existing techniques.