OEE (Overall Equipment Effectiveness)

OEE answers one question: of all the good fabric this loom could theoretically have produced during the time you planned to run it, what fraction did you actually get? An OEE of 100 percent would mean the machine ran every planned minute, at its rated speed, producing nothing but first-quality output. No mill achieves that, and the value of the metric lies precisely in where it falls short.

The OEE formula

OEE is the product of three factors, each expressed as a percentage:

Availability is the share of planned production time during which the machine was actually running. It is reduced by breakdowns, beam changeovers, waiting for yarn, and unplanned stoppages. Performance is the share of the rated speed the machine achieved while it was running, reduced by slow running and micro-stops. Quality is the share of output that met specification on the first pass, reduced by defects and rework.

OEE = Availability x Performance x Quality. Because the three multiply rather than average, OEE punishes weakness anywhere in the chain. Three respectable-looking factors of 90 percent each produce an OEE of only 73 percent.

A worked example on a loom

Take a weaving shed running a single shift of 8 hours, or 480 planned minutes. During that shift the loom is stopped for 40 minutes for a beam change and 20 minutes for a warp break, leaving 420 minutes of actual run time.

The loom looks busy. An owner walking the floor would see it running for most of the shift. Yet a quarter of its capacity evaporated, and no single number on the shop-floor register would have revealed that. This is the entire argument for OEE: it makes invisible losses arithmetic.

Realistic benchmarks for textile mills

The frequently quoted figure of 85 percent as world-class OEE originates in discrete manufacturing and transfers to textiles only loosely. Weaving, knitting, dyeing and finishing have very different changeover profiles, and a dye house that runs long batches will post a higher OEE than a weaving shed running short runs of varied constructions, without being better managed.

In practice most textile mills measure between 55 and 70 percent once they begin recording honestly. Top-quartile mills reach the low 80s. A mill that computes 90 percent in its first month of measurement has almost always made a definitional error, usually by excluding changeovers from planned time.

That exclusion is the most common way OEE gets quietly gamed. If beam changes are treated as planned downtime and removed from the denominator, availability rises, OEE rises, and the mill has learned nothing — because changeover time is exactly the loss it most needs to attack. Include everything you intended to produce during, and let the number be uncomfortable.

What OEE does not tell you

OEE measures how well you ran the machines you chose to run. It is silent on whether you should have been running them at all. A loom producing first-quality fabric at 92 percent OEE against an order nobody wants is destroying value efficiently, and OEE will applaud it.

It also aggregates badly. A shed-wide OEE of 70 percent may conceal four looms at 85 percent and one at 20 percent, and the average invites you to launch an improvement programme across all five when one machine needs a bearing. OEE is diagnostic at machine level and merely descriptive above it.

Turning the number into capacity

The reason OEE justifies attention is that its losses are usually cheaper to recover than to buy. Lifting a shed from 65 to 80 percent OEE releases roughly 23 percent more output from machines already owned, paid for and installed. Against the capital cost and lead time of new looms, that is an unusually good trade — provided the losses are attacked in the right order.

The order is dictated by the arithmetic. Because the factors multiply, the largest single deficit yields the largest gain. A mill at 87.5 percent availability, 90 percent performance and 95.2 percent quality should attack changeover and breakdown time first, not chase the last 5 percent of defects, even though defects feel more visible and generate more complaints.

Vastra ERP records availability, performance and quality losses separately at machine level across looms, dyeing machines and finishing lines, so a mill can see which of the three factors is costing it capacity on which machine, rather than watching a blended shed average that hides the answer.