Overall Equipment Effectiveness, OEE
OEE is a composite metric of how productively a piece of manufacturing equipment is operating. It multiplies three factors. Availability, fraction of scheduled time the equipment was running, performance, rate vs design, and quality, fraction of output meeting spec. OEE on packaging lines clusters around 85% at the top end. Typical mid-tier plants run 50-70%.
Read one of your own drawings.
Drop a P&ID, instrument index, or schedule. Tagsight reads it to the tag and opens a workspace you keep when you sign in.
PDF · DWG · DXF · TIFF · PNG · XLSX
Overall equipment effectiveness is the single number that asks how much useful output a piece of equipment produced against what it could have produced if it ran all its scheduled time, at full rate, with no rejects. It is the product of three fractions. Availability, the share of planned time the equipment was actually running. Performance, the actual rate against the design rate. And quality, the share of output that met specification. Because it is a product, a high score requires all three to be high at once, and a plant with strong availability and rate but a quality problem still scores poorly, which is the point. OEE forces the three loss categories into one comparable figure. Its most honest use is as a trend on the same equipment over time rather than a benchmark between dissimilar lines, because the bottleneck and the rate basis differ from one process to another, and a packaging line and a continuous reactor are not measuring the same thing when each reports a number. It also says nothing about the cost of the losses, so a disciplined improvement program reads OEE alongside a loss tree that values each loss in money rather than chasing the figure for its own sake. The data behind it, run time, output, reject count is reconciled in the MES, which is itself fed from the controller dataset the I/O list defines, and a plant computing the metric from the historian and operator logs without an MES typically waits days rather than minutes for a clean number.
How OEE is calculated.
Availability, run time, planned production time. Performance, actual output * ideal cycle time, run time. Quality, good output, total output. OEE Availability * Performance * Quality. Each factor is a percentage. The product is a percentage. A line with 90% availability, 85% performance, 95% quality has OEE 0.90 * 0.85 * 0.95 0.727 or 72.7%.
OEE pitfalls.
OEE is most useful as a trend over time on the same equipment, not as a benchmark across different lines. Comparing OEE across a packaging line and a process reactor is meaningless because the bottleneck definitions differ. OEE also doesn't capture the cost of the losses. An OEE-improvement project that cuts changeover time by 20% may matter less than one that reduces unplanned downtime by 5% in dollar terms. Use OEE alongside loss-tree analysis rather than as a single number.
Frequently asked.
Where does OEE data come from.
The MES, where production-order, run-time, and output data are reconciled. Some plants compute it directly from the historian and operator logs without an MES, but the cycle to produce a clean OEE number jumps from minutes to days.
Is OEE a process-industry metric or a discrete-manufacturing metric.
Both, but it originated in discrete manufacturing where the cycle-time-per-piece concept is straightforward. Process plants adapt OEE to continuous operations by defining performance against a design-rate basis rather than a per-piece cycle time.
Who is responsible for improving OEE on a process line.
No single role owns it entirely. Reliability engineers target availability through predictive maintenance programs. Process engineers and BPCS engineers target performance by optimizing control-loop tuning and minimizing grade transitions. Quality teams target the quality factor. In practice, a cross-functional loss-tree review identifies which factor is limiting, and the relevant discipline leads the improvement project.