Factory-level performance evaluation of buffered multi-state production systems

Abstract

This paper proposes a performance evaluation model for production systems considering: i) the need of a factory-level performance metric tracking system effectiveness and ii) the need of ex-ante performance evaluation models for buffered multi-state systems. Potentialities of Overall Equipment Effectiveness (OEE) calculation approach and reliability analysis methods are combined. Indeed, even if OEE is widely adopted in the industrial practice, including different inefficiency factors, its use for supporting asset-related decision-making has got two main pitfalls. Firstly, it is limited to equipment level without considering the typical complexity of production systems. Secondly, it is traditionally computed through ex-post analysis of data, being not sufficient for decision-making at the beginning of the asset lifecycle. Conversely, in the reliability theory, different modelling methods are proposed for multi-state systems performance evaluation. Nevertheless, they usually focus on reliability/availability computation, lacking specific consideration of buffers and not measuring the inefficiency factors that may affect a system, relevant for practitioners. The combination of the two approaches is proposed, with the purpose to develop a simulation-based performance evaluation model for buffered multi-state production systems and to compute the Overall Factory Effectiveness (OFE). The model is applied in a manufacturing company to support evaluating improvement scenarios of an existing plant.