One number that tells you how much capacity is actually showing up.
Overall equipment effectiveness is the most useful number a small shop can put on a board, and also the most easily turned into theater. The strength of OEE is that it captures three different ways a machine can lose capacity in one figure that you can plot over time. The weakness is that without the underlying loss data, the number is just a score, and a score that nobody can act on becomes a ritual. The shops that use OEE well treat the number as a starting point for asking which of the three factors is hurting today.
"OEE on a board is a score. OEE with the losses underneath is a plan."
OEE multiplies three percentages. Availability is the run time divided by the planned production time. If a shift is eight hours and the machine ran for seven, availability is 87.5 percent. Performance is the actual output divided by the theoretical output at the machine's ideal cycle time. If the machine should have run 700 parts in that seven hours and only ran 630, performance is 90 percent. Quality is the good parts divided by the total parts. If 600 of those 630 passed inspection, quality is 95.2 percent. The OEE is 0.875 times 0.9 times 0.952, or about 75 percent.
The three factors are independent levers. A machine can have high availability but low performance because it runs slowly between breakdowns. It can have great performance but poor quality because it runs fast and scraps a lot. Or it can have all three issues at once, and the OEE will reveal that there is no single bottleneck. The diagnostic structure underneath OEE is the six big losses, which assigns each loss type to one of the three factors so improvement work has a target.
The benchmark numbers most shops hear (85 percent is world class, 60 percent is the typical starting point for an SMB) are loose guidelines, not standards. The right comparison is the same machine to itself over a quarter, not your shop to someone else's. A 65 percent OEE that improves to 72 percent over two months is doing more for the business than a fixed 85 percent that does not move.
Picture a 25 person machine shop running a Mazak with two operators on day shift and one on swing. The owner suspects the machine is the bottleneck but cannot prove it. A six week OEE study reveals: availability 78 percent (mostly from changeover and one chronic spindle issue), performance 86 percent (from minor stops and conservative feed rates the operators use to avoid scrap), quality 96 percent. The OEE: 64 percent.
The improvement targets are now clear. Spindle issue gets escalated to the maintenance lead, which adds about three percentage points to availability over a month. Operators are trained on feed rates that the program can actually handle, adding four points to performance. After a focused improvement effort, OEE on the Mazak climbs from 64 to 72 percent. That is roughly 13 percent more capacity from the same machine, same staff, same shift. The shop did not need a new spindle; it needed a number on the board and a list of losses underneath.
OEE is built from availability, performance rate, and quality rate, three factors that decompose every equipment loss into one of three modes. The diagnostic framework that maps every type of loss to one of those factors is the six big losses. OEE is the canonical scoreboard for any total productive maintenance program.
The questions we hear most about this term.
Long-form guides that pick up where this definition leaves off, written for manufacturers running Arda today.
Same-day setup. No distributor lock-in. Zero stockouts. Top teams double revenue in 9 months.
.svg)
.svg)
%201.svg){kind=logo}
.png)
.svg)
.svg)
