SPCVisualsSEMI E10 availability, MTBF and E79 OEE — charted, not compared
SEMI E10 splits a tool's time into six states — productive, standby, engineering, scheduled downtime, unscheduled downtime and non-scheduled — and defines availability, utilisation, MTBF and MTTR from them. SEMI E79 multiplies four time ratios from the same states into overall equipment efficiency.
One line per month with the hours in each state and the failure count. You get every E10 metric per period and for the whole window, OEE if you give the rated throughput, and the metric you choose drawn on an individuals chart with limits — so the question "did reliability actually change?" gets an answer instead of an argument.
How you calculate E10 availability and OEE in Excel
One column per state, one row per month. Availability is =SUM(B2:D2)/(SUM(B2:G2)−G2) with the six states in B to G; MTBF is productive hours divided by the failures column; MTTR is unscheduled hours divided by the same; E79 OEE is (good units / rated throughput) / SUM(B2:G2). AVERAGE down each column gives the figure for the report.
The report then puts this month beside last month, and that comparison has no yardstick. Availability of 87% after 89% is either a tool getting worse or a tool doing what it always does, and nothing on the sheet says which. The moving range between consecutive months is the yardstick — it is the same arithmetic as the chart below, and Excel will not draw it for you without a dozen helper cells.
Nothing you paste leaves this calculation. It is not stored, not logged and not sent anywhere else — the numbers are computed and returned, and that is all. Paste straight out of a spreadsheet — tab-separated columns are understood as they land.
| Period | Availability | Op. utilisation | MTBF h | MTTR h | Failures | OEE |
|---|
The six states, and what E10 builds from them
SEMI E10 is a bookkeeping standard before it is anything else. Every hour of a tool's existence goes into exactly one of six states, and every reliability metric is a ratio of those buckets:
| State | What it means | Counts as |
|---|---|---|
| Productive | Making product, including rework and engineering lots run as production | Uptime |
| Standby | Able to run but idle: no lots, no operator, waiting on upstream | Uptime |
| Engineering | Running for process or equipment engineering, not production | Uptime |
| Scheduled downtime | Planned maintenance, setup, qualification, change of consumables | Downtime |
| Unscheduled downtime | Repair after a failure, including the wait for parts and people | Downtime |
| Non-scheduled | Not expected to run: no shift staffed, shutdown, tool not yet installed | Outside operations time |
Then: operations time is total time minus non-scheduled time. Availability is uptime over operations time. Total utilisation is productive time over total time, and operational utilisation is productive time over operations time. MTBF is productive hours per failure and MTTR is unscheduled downtime hours per failure. Nothing here needs more than a division, and the whole difficulty of E10 in practice is getting the hours into the right bucket in the first place — in particular, agreeing what counts as a failure versus an assist, which E10 also defines and which no calculator can decide for you.
E79: the four factors of OEE
SEMI E79 defines overall equipment efficiency for semiconductor tools as a product of four efficiencies, each a ratio of times built from the same E10 states plus a rated throughput:
- Availability efficiency — uptime over total time. Note the denominator: E79 uses total time here, where E10 availability uses operations time.
- Operational efficiency — productive time over uptime. What fraction of the time it could have run, it did.
- Rate efficiency — the time the actual output should have taken at rated speed, over the productive time it actually took.
- Quality efficiency — the theoretical time for the good units over the theoretical time for all units, which reduces to good units over units.
Multiply them and everything cancels: OEE is the hours the good output should have taken divided by the hours the tool was there for. That is the number to check the four factors against, and it is why a small tool can compute E79 correctly with one extra input — the rated throughput — and nothing else.
Why the chart, and why an individuals chart
Every metric above is reported monthly, and reported monthly it is compared with last month. That comparison is the thing this page exists to argue against. A tool's availability moves from month to month whether or not anything changed: a different mix of lots, one long repair instead of three short ones, a month with 672 hours in it instead of 744. Two figures side by side cannot tell that movement from a real shift, so the meeting explains whichever way it went.
The individuals chart uses the movement between consecutive months to draw the band a stable tool wanders within — limits at the mean ± 2.66 times the mean moving range. A month outside the band is a signal worth a root cause. A month inside it, however much worse than the last, is the tool doing what it does, and the honest answer to "why is availability down?" is "it is not, measurably". That answer is unavailable without the chart.
Rules 1 to 4 are evaluated here — beyond the limits, nine on one side, six trending, fourteen alternating — and the zone rules are not, because a dozen bounded percentages is not the sample the zone rules assume. The I-MR chart page covers the chart itself in more depth, including the one assumption it makes that equipment data can violate: consecutive months are treated as independent, and a repair programme spread over a quarter makes them not.
What this calculator does not do
- Decide what a failure is. E10 distinguishes failures from assists and from interrupts by whether the tool could continue without intervention. That classification happens on the floor, and the MTBF is only as honest as it is.
- Equipment-dependent versus supplier-dependent uptime. E10 also splits downtime by who was responsible. It needs a cause per event and is out of scope here.
- The cumulative E10 metrics — MTBA, MCBF, MTOL. Assists and cycles are not in the paste. The arithmetic is the same shape; the columns are not.
- Watch the tool. The chart is drawn from what you pasted, once. A chart that is fed every shift and raises the flag itself is the product, and the E10 standards page says where it does and does not belong in a fab.
What the Excel version will not do
- Averaging percentages weights a short month like a long one
- February at 672 hours and March at 744 do not carry equal weight in a window figure. The window number is the ratio of the summed hours, and AVERAGE of the monthly ratios is a different, wrong number that nobody in the meeting will notice.
- Zero failures becomes a divide-by-zero, or an infinite MTBF
- A quiet month either errors the cell or inflates the average with a nonsense figure. MTBF is undefined for that month, and the honest sheet leaves it blank — which the average then silently skips.
- The sheet is opened once a month, after the fact
- Equipment metrics are reviewed at the monthly meeting, which is the worst possible cadence for a signal that happened in week one. A chart that recomputes as the hours land and raises a flag the day a limit is crossed is the difference between a report and a control.
Questions people ask about this
What are the six SEMI E10 equipment states?
Productive (making product), standby (able to run but idle — no operator, no lots, no material), engineering (running for process or equipment engineering, not production), scheduled downtime (planned maintenance, setup, qualification), unscheduled downtime (repair after a failure, waiting for parts or a technician) and non-scheduled time (the tool is not expected to run — a holiday shutdown, no shift staffed). The first three are uptime, the next two are downtime, and the last is outside operations time altogether.
How is availability calculated under SEMI E10?
Equipment uptime divided by operations time, as a percentage. Uptime is productive plus standby plus engineering hours; operations time is total time minus non-scheduled time. Excluding non-scheduled time is the E10 choice that matters: a tool that sat idle over a plant shutdown is not penalised for hours it was never asked to run.
How are MTBF and MTTR calculated for semiconductor equipment?
E10 uses productive-time MTBF: productive hours divided by the number of failures in the period. MTTR is unscheduled downtime hours divided by the same failure count. A period with no failures has no MTBF — the interval is open-ended, not infinite — which is why this calculator reports it as undefined rather than as a very large number.
What is the difference between E10 availability and E79 OEE?
Availability asks whether the tool could run. OEE asks how much good product came out relative to what the tool could have made if it had run at rated speed, all the time, with no scrap. E79 factors it as availability efficiency × operational efficiency × rate efficiency × quality efficiency, and every factor is a ratio of times, so it collapses to the hours the good output should have taken divided by the hours the tool was there for.
Why chart availability instead of comparing this month with last month?
Because a monthly figure moves every month whether or not anything changed. Two numbers give you no way to tell a real shift from normal variation, and the meeting ends up explaining noise. An individuals chart with limits from the moving range draws the band a stable tool wanders within; a month outside it is a signal, and a month inside it, however much lower than last month, is not.
One tool, one chart, watched every shift — free.
Equipment availability is one characteristic per tool, which is exactly the shape of the free plan. Push the state hours by API from the MES or type them in at the end of the shift, freeze the limits on a baseline year, and the maintenance manager gets a signal the day it happens instead of a comparison with last month at the end of the next one.
Other free tools
- Cp / Cpk calculator — Paste measurements — or type a mean and a sigma — with your tolerance, and get Cp, Cpk, Pp, Ppk, the sigma level and the expected parts per million out of spec.
- Control limit calculator — Give it subgroups or individual readings and it returns the control limits for the chart, the range or sigma chart beneath it, and every constant it used to get there.
- Control chart generator — Paste a column of numbers, or rows of subgroups, and get a real control chart: limits from the data, Nelson rules 1–4 evaluated, out-of-control points marked.
- Nelson & Western Electric rules checker — Every run rule evaluated on your data, each violation named in plain English with what it usually indicates — a shift, a trend, tool wear, two machines mixed.
- Cpk confidence interval & sample size — A Cpk of 1.33 from 30 pieces has a 95% interval of roughly 0.97 to 1.69. See the uncertainty in your own number, and how many parts would settle it.
- Cpk ↔ PPM ↔ sigma level converter — What PPM is a Cpk of 1.33? What Cpk does 3.4 PPM need? Both conventions shown side by side, because the 1.5 sigma shift is why two sources disagree by a factor of ten.
- Cm / Cmk machine capability calculator — Paste the parts from a machine buy-off with your tolerance and get Cm and Cmk — the AIAG-VDA manual's Pm and Pmk — with the spread convention stated and the run order charted.
- Control chart constants — The whole table, n = 2 to 25, with the formula each constant belongs to. The same values the charts on this site are computed from.
- X̄-R chart calculator — Subgroups in, X̄ and R charts out — limits from A₂, D₃ and D₄, run rules evaluated, and the range chart shown first because it decides whether the averages chart can be trusted.
- X̄-s chart calculator — For subgroups big enough that the range wastes them. Limits from A₃, B₃ and B₄, sigma recovered with the c₄ correction, and the s chart read first because it decides whether the averages chart can be trusted.
- I-MR chart calculator — For processes that give you one number at a time — a batch, an oven, a destructive test. Limits from the mean moving range, run rules evaluated, and an honest note about what an individuals chart cannot see.
- Levey-Jennings chart with Westgard rules — Paste your QC results with the assigned mean and SD from the package insert, and get the chart with every sigma band drawn and all six Westgard rules evaluated — each one named, and labelled random or systematic error.
- Bland-Altman plot — Paired readings from two methods, plotted as difference against average — with the bias, the limits of agreement, their confidence intervals, and a test for whether the disagreement depends on the magnitude.
- Pareto chart generator — Categories and counts in, ranked bars and the cumulative line out — with the vital few named, an "Other" bucket that always sorts last, and an honest warning when the distribution is flat and there is no dominant cause to attack.
- Box and whisker plot generator — One column per group, boxes side by side. Quartiles, the 1.5×IQR fences, whiskers that stop at real readings and outliers drawn individually — with the quartile method stated, because that is why your plot and Excel's disagree.
- Histogram and normality test — A histogram with the fitted normal curve, both standard bin rules with a reasoned recommendation, and an Anderson-Darling test that refuses to tell you your data is normal — because no test can.
- Gage R&R calculator (ANOVA) — Paste a crossed study — part, operator, reading — and get the full ANOVA: repeatability and reproducibility separated, the part-by-operator interaction tested rather than assumed away, %GRR, %Tolerance and ndc against AIAG's bands.
- p, np, u and c chart calculator — Counted data rather than measured: proportion defective, number defective, defects per unit. Limits that step with the sample size instead of pretending it never changed, and an honest warning when the counts are too low for three-sigma limits to mean anything.
- Grubbs outlier test — The G statistic against its critical value, the p-value, and the point implicated — plus a straight answer about what to do next, which is almost never to delete the reading.
- Scatter diagram and correlation — Paired readings plotted, with Pearson r, Spearman rho, r², the least-squares line and its p-value — and the two reported together, because when they disagree the disagreement is the finding.
- Standard deviation calculator — The usual summary statistics, and the thing a spreadsheet will not tell you: the within-subgroup sigma a control chart uses, beside the overall sigma a textbook means. The gap between them is why Cpk and Ppk disagree.