DPMO Calculator

Q: What is a good DPMO score?

It depends on the industry, but here is a quick guide: Under 233 DPMO (5σ or better): excellent, used in aerospace and medical devices. 233 to 6,210 DPMO (4σ to 5σ): above average, typical of strong manufacturing lines. 6,210 to 66,807 DPMO (3σ to 4σ): average, where most companies sit. Over 66,807 DPMO (below 3σ): poor, and the process needs real fixing. Safety-critical work should aim much lower than a typical factory line.

Q: What is the difference between a defect and a defective unit?

A defect is one thing that is wrong. A defective unit is any unit with at least one defect. Example: one circuit board with 3 bad solder joints counts as 3 defects but only 1 defective unit . DPMO counts defects. Percent defective counts units. That is why the two numbers never match.

Q: How do you convert DPMO to sigma level in Excel?

Use the inverse normal function and add the 1.5 shift: =NORM.S.INV(1 - (DPMO/1000000)) + 1.5 So for 6,210 DPMO you get about 4.0 sigma . To go the other way, from sigma to DPMO, use: =(1 - NORM.S.DIST(sigma - 1.5, TRUE)) * 1000000

Q: What is the difference between DPMO and PPM?

PPM (parts per million) counts bad parts out of a million parts. It ignores how many ways a part can fail. DPMO counts bad opportunities out of a million chances to fail. If each unit has only 1 opportunity, the two numbers are the same. If a unit has 10 opportunities, DPMO will be about 10 times lower than PPM. Always say which one you are using.

Q: What sigma level is 99% yield?

A 99% yield is about 3.83 sigma , which equals 10,000 DPMO . That sounds strong, but it means 1 in every 100 chances fails. For comparison, 4 sigma is 99.38% yield and 6 sigma is 99.99966% yield. Small jumps in yield near the top mean big jumps in sigma.

Q: What is the difference between first pass yield and rolled throughput yield?

First pass yield (FPY) is the percent of units that pass one step with no rework. Rolled throughput yield (RTY) multiplies the yields of all steps together. Example: four steps at 95% each. Each FPY is 95%, but RTY = 0.95 × 0.95 × 0.95 × 0.95 = 81.5% . RTY shows the hidden factory that single-step numbers hide.

DPMO Calculator

Introduction

DPMO stands for Defects Per Million Opportunities. It tells you how many defects you would find if your process ran one million chances to fail. It is one of the main ways to measure quality in Six Sigma.

This calculator converts shop floor counts into quality metrics. Enter the defects you found, the units you checked, and how many ways each unit can fail. You get DPMO, sigma level, yield, DPO, DPU, and rolled throughput yield (RTY).

You can also work backward. Six modes let you convert DPMO to sigma, yield to sigma, sigma to DPMO, or DPU to DPMO. You can even compare two lines side by side to see which one runs better.

Every answer comes with step-by-step math, a chart, and a quality rating from Poor to World Class. Use it to track a line, set goals, or check if a fix really worked.

How to use our DPMO Calculator

Pick a calculation mode and type in your process numbers. The calculator returns DPMO, sigma level, yield, DPO, and a quality rating, along with step-by-step math and a chart.

Calculation Mode: Choose what you want to find. You can go from defects to DPMO, from DPMO to sigma, from yield to sigma, from sigma to metrics, from DPU to DPMO, or compare two processes. The input boxes change to match your pick.

DPMO (Mode 1): Type your known defects per million opportunities, from 0 to 1,000,000. The tool turns it into sigma level, yield, and DPO.

Defects Found (Mode 2): Enter the total number of defects you counted on all the units you checked. This number cannot be negative or larger than your total opportunities.

Units Inspected (Mode 2): Enter how many units you looked at. This is optional, but adding it unlocks DPU, DPM, rolled throughput yield, and percent defective units.

Opportunities Per Unit (Mode 2): Enter how many ways one unit can have a defect, such as solder joints or checks. It must be 1 or more. This box is also optional.

Total Opportunities (Mode 2): Enter the total chances for a defect. If you filled in Units and Opportunities Per Unit, the calculator fills this in for you by multiplying them.

Process Yield % (Mode 3): Enter the percent of opportunities that came out with no defect, from 0 to 100. You get the matching sigma level and DPMO.

Sigma Level (Mode 4): Enter a target sigma level from 0 to 6. The tool shows the DPMO and yield that level needs, using the standard 1.5 sigma shift.

DPU (Mode 5): Enter the average number of defects on one unit. It cannot be more than the opportunities per unit.

Opportunities Per Unit (Mode 5): Enter the number of defect chances on each unit. It must be 1 or more, and it turns your DPU into DPMO.

Process A and B Labels (Mode 6): Type a name for each process, like a line or shift name. Names are optional and only show on the result cards.

Process A and B DPMO (Mode 6): Enter the DPMO for each process, from 0 to 1,000,000. The calculator names the winner and shows how much better it is.

Calculate, Try Example, and Clear buttons: Press Calculate to see your results. Press Try Example to load sample numbers, or Clear to empty the boxes.

Interactive Sigma Visualizer sliders: Drag the Defects, Units, and Opportunities Per Unit sliders to watch DPMO, sigma level, yield, and DPU change live on the sigma scale.

What Is DPMO?

DPMO stands for Defects Per Million Opportunities. It tells you how many defects a process would make if it ran one million chances to fail. It is one of the main quality measures used in Six Sigma and in manufacturing quality control.

The DPMO Formula

DPMO uses three numbers: defects found, units checked, and opportunities per unit.

Total Opportunities = Units × Opportunities Per Unit
DPO = Defects ÷ Total Opportunities
DPMO = DPO × 1,000,000

Example: 27 defects in 1,500 units, with 9 chances to fail on each unit. That is 13,500 opportunities. 27 ÷ 13,500 = 0.002, and 0.002 × 1,000,000 = 2,000 DPMO.

What Is an "Opportunity"?

An opportunity is any single way one unit can go wrong. A circuit board with 9 solder joints has 9 opportunities. A form with 12 fields has 12 opportunities. Counting opportunities the same way every time matters. If you change the count, the DPMO changes too, and you cannot compare old data to new data.

Sigma Level and Yield

DPMO can be turned into a sigma level, a score from 0 to 6 that shows how good a process is. Higher is better. Most companies add a 1.5 sigma shift to the math. This shift allows for the fact that real processes drift over time and are not perfectly steady. Yield is the flip side of DPMO, the percent of chances that came out with no defect.

Sigma Level Chart

Sigma Level	DPMO	Yield	Rating
6.0σ	3.4	99.99966%	World Class
5.0σ	233	99.977%	Excellent
4.0σ	6,210	99.379%	Above Average
3.0σ	66,807	93.32%	Average
2.0σ	308,538	69.15%	Below Average
1.0σ	691,462	30.85%	Poor

Other Metrics You Will See

DPU (Defects Per Unit): defects ÷ units. It ignores opportunities, so it only works when comparing the same kind of product.
DPO (Defects Per Opportunity): the plain decimal before you scale it to a million.
DPM (Defects Per Million Units): DPU scaled to a million units.
RTY (Rolled Throughput Yield): the percent of units that pass every step with zero defects.

Why DPMO Matters

DPMO lets you compare very different processes on one fair scale. A car assembly line and a hospital billing desk do not share units, but both can be scored in DPMO. Teams use it to set goals, track progress after a fix, and decide which process needs work first. Moving from 3 sigma to 4 sigma cuts defects by more than 90%, so even small sigma gains save real money.

Things to Watch For

DPMO counts every defect, even two on the same unit. That is why DPMO and percent-defective units are not the same number. Also, a defect must be clearly defined before you count. If two inspectors judge the same part in different ways, your DPMO is not reliable. Use a big enough sample too. A handful of units will not give a trustworthy sigma level.

Formulas used

Defects Per Opportunity (DPO)

\text{DPO} = \frac{D}{N_o} = \frac{D}{U \times O}

Defects Per Million Opportunities (DPMO)

\text{DPMO} = \text{DPO} \times 1{,}000{,}000

Process Yield

Y = (1 - \text{DPO}) \times 100\%

Sigma Level from DPO (with 1.5 sigma shift)

\sigma = Z(1 - \text{DPO}) + 1.5 = \Phi^{-1}(1 - \text{DPO}) + 1.5

DPMO from Sigma Level

\text{DPMO} = \left[1 - \Phi(\sigma - 1.5)\right] \times 1{,}000{,}000

Defects Per Unit (DPU) and Defects Per Million Units (DPM)

\text{DPU} = \frac{D}{U}, \qquad \text{DPM} = \text{DPU} \times 1{,}000{,}000

Rolled Throughput Yield and Percent Defective Units

\text{RTY} = (1 - \text{DPO})^{O}, \qquad \%\,\text{defective} = (1 - \text{RTY}) \times 100\%

Process Comparison Improvement

\text{Improvement} = \frac{\text{DPMO}_{\max} - \text{DPMO}_{\min}}{\text{DPMO}_{\max}} \times 100\%

Frequently asked questions

What is a good DPMO score?

It depends on the industry, but here is a quick guide:

Under 233 DPMO (5σ or better): excellent, used in aerospace and medical devices.
233 to 6,210 DPMO (4σ to 5σ): above average, typical of strong manufacturing lines.
6,210 to 66,807 DPMO (3σ to 4σ): average, where most companies sit.
Over 66,807 DPMO (below 3σ): poor, and the process needs real fixing.

Safety-critical work should aim much lower than a typical factory line.

Why is Six Sigma 3.4 defects per million and not 2 per billion?

Pure math says a process running 6 sigma from the target should make about 2 defects per billion. But real processes drift. Tools wear, materials change, and operators change.

Six Sigma adds a 1.5 sigma shift to allow for that drift over time. So 6 sigma minus 1.5 equals 4.5 sigma of real performance, which works out to 3.4 defects per million opportunities.

What is the difference between a defect and a defective unit?

A defect is one thing that is wrong. A defective unit is any unit with at least one defect.

Example: one circuit board with 3 bad solder joints counts as 3 defects but only 1 defective unit.

DPMO counts defects. Percent defective counts units. That is why the two numbers never match.

How do you convert DPMO to sigma level in Excel?

Use the inverse normal function and add the 1.5 shift:

=NORM.S.INV(1 - (DPMO/1000000)) + 1.5

So for 6,210 DPMO you get about 4.0 sigma. To go the other way, from sigma to DPMO, use:

=(1 - NORM.S.DIST(sigma - 1.5, TRUE)) * 1000000

What is the difference between DPMO and PPM?

PPM (parts per million) counts bad parts out of a million parts. It ignores how many ways a part can fail.

DPMO counts bad opportunities out of a million chances to fail.

If each unit has only 1 opportunity, the two numbers are the same. If a unit has 10 opportunities, DPMO will be about 10 times lower than PPM. Always say which one you are using.

Can DPMO be higher than 1,000,000?

No. Each opportunity can either pass or fail, so the worst case is every opportunity failing, which is exactly 1,000,000 DPMO.

If your math gives more than that, you either counted defects that are not tied to real opportunities, or your opportunity count per unit is too low. Recheck how you defined an opportunity.

Does counting more opportunities per unit lower your DPMO?

Yes, and that is a common trap. If you split one check into five checks, the defect count stays the same but the opportunity count jumps, so DPMO drops and sigma looks better.

Nothing actually improved. Lock down your opportunity definition in writing, count only real, independent ways to fail, and never change it mid-project.

What sigma level is 99% yield?

A 99% yield is about 3.83 sigma, which equals 10,000 DPMO.

That sounds strong, but it means 1 in every 100 chances fails. For comparison, 4 sigma is 99.38% yield and 6 sigma is 99.99966% yield. Small jumps in yield near the top mean big jumps in sigma.

What is the difference between first pass yield and rolled throughput yield?

First pass yield (FPY) is the percent of units that pass one step with no rework.

Rolled throughput yield (RTY) multiplies the yields of all steps together.

Example: four steps at 95% each. Each FPY is 95%, but RTY = 0.95 × 0.95 × 0.95 × 0.95 = 81.5%. RTY shows the hidden factory that single-step numbers hide.

How many units do you need to inspect for a reliable DPMO?

Enough to expect at least 5 to 10 defects. With zero or one defect, your DPMO is mostly luck.

Rough rule: divide 5 by your expected DPO to get the opportunities you need. If you expect 1 defect per 1,000 opportunities, check about 5,000 opportunities. Very low defect rates need huge samples, which is why high-sigma processes are hard to measure by inspection alone.

How do you lower DPMO in a process?

Attack the causes, not the symptoms:

Find the vital few: sort defects by type and fix the top one or two first.
Mistake-proof it: use fixtures, sensors, and checks that make the error impossible.
Cut variation: control temperature, torque, speed, and material lots.
Standardize work: same method, same result, every shift.
Verify the fix: re-measure DPMO after the change, not before.

What is the difference between short-term and long-term sigma?

Short-term sigma is how the process performs in a short, steady run with no drift. Long-term sigma is real performance over weeks or months, including shift changes and tool wear.

The usual link is: short-term = long-term + 1.5. When someone says a process is 6 sigma, they mean short-term, which is really 4.5 sigma over the long haul.

How is DPMO used outside of manufacturing?

Any process with countable errors can use DPMO. Examples:

Hospital billing: each invoice field is an opportunity, wrong codes are defects.
Call centers: each call has checks like greeting, ID check, and notes.
Software: each test case or field entry is an opportunity.
Shipping: each order line, label, and address counts.

Because it scales to a million chances, DPMO lets you compare an office process to a factory line on one fair scale.

Rate this calculator