Example: Cpk for a Sheet Metal Bracket Hole Diameter
Scenario: You're manufacturing a sheet metal bracket with a critical hole that needs to be within a tight
tolerance for proper assembly. Let's say the target diameter of the hole is 10 mm, with an upper
specification limit (USL) of 10.1 mm and a lower specification limit (LSL) of 9.9 mm. You collect a
sample of 25 hole diameter measurements during production.
1. Data Collection:
You measure the diameter of 25 brackets and record the values (data can be hypothetical for
this example).
2. Data Visualization (Optional):
Create a histogram of the collected diameter measurements. This will show the distribution of
the data points and any potential issues like skewness.
3. Define Specification Limits:
USL = 10.1 mm
LSL = 9.9 mm
4. Calculate Process Capability Indices:
Spec Width (USL - LSL) = 10.1 mm - 9.9 mm = 0.2 mm
You'll need to calculate the standard deviation (σ) from your sample data using statistical
methods or software. Suppose the standard deviation is found to be 0.03 mm.
5. Calculate Cpk:
Cpk = min { (USL - μ) / (3σ), (μ - LSL) / (3σ) }
Let's say the average (mean, μ) of your diameter measurements is 10.02 mm.
Cpk = min { (10.1 - 10.02) / (3 * 0.03), (10.02 - 9.9) / (3 * 0.03) }
Cpk = min { 0.027 / 0.09, 0.04 / 0.09 }
Cpk = 0.3 (assuming this is the minimum value)
6. Interpretation:
A Cpk of 0.3 indicates the process has low capability. There's a high chance of parts falling
outside the specification limits (9.9 mm - 10.1 mm).
You might need to investigate the cause of the variation and improve the manufacturing
process to achieve a higher Cpk value (ideally > 1.33 for better quality control).
Note: This is a simplified example. In real-world scenarios, you'll use the actual data from your
measurements to calculate Cpk.
Here's how to calculate the standard deviation (σ) using statistical methods for the sheet metal bracket
hole diameter example:
Method 1: Using the Population Standard Deviation Formula (if data represents the entire
population)
If your data represents the entire population of hole diameters being produced (i.e., you measured every
single bracket), then you can use the population standard deviation formula:
σ = √( Σ(x - μ)^2 / N )
where:
σ (sigma) is the population standard deviation
Σ (sigma) represents the sum of
x is each individual hole diameter measurement
μ (mu) is the population mean (average) of all hole diameter measurements
N is the total number of measurements in the population (all brackets)
Steps:
1. Calculate the mean (μ) of all your diameter measurements by adding them up and dividing by
the total number (N) of measurements.
2. For each measurement (x), subtract the mean (μ) to find the deviation from the mean. Square
each deviation (x - μ)^2.
3. Add up the squared deviations from all measurements ( Σ(x - μ)^2 ).
4. Divide the sum of squared deviations by the total number of measurements (N).
5. Take the square root of the result obtained in step 4. This will be your population standard
deviation (σ).
Method 2: Using the Sample Standard Deviation Formula (if data represents a sample)
More commonly, you'll likely be working with a sample of measurements that represents the entire
production process. In this case, use the sample standard deviation formula:
s = √( Σ(x - x̄ )^2 / (n - 1) )
where:
s is the sample standard deviation
Σ (sigma) represents the sum of
x is each individual hole diameter measurement
x̄ (x bar) is the sample mean (average) of the measurements in the sample
n is the total number of measurements in the sample
Steps:
1. Calculate the sample mean (x̄ ) of your diameter measurements by adding them up and
dividing by the number (n) of measurements in the sample.
2. For each measurement (x), subtract the sample mean (x̄ ) to find the deviation from the mean.
Square each deviation (x - x̄ )^2.
3. Add up the squared deviations from all measurements in the sample ( Σ(x - x̄ )^2 ).
4. Divide the sum of squared deviations by the number of measurements in the sample minus
one (n - 1).
5. Take the square root of the result obtained in step 4. This will be your sample standard
deviation (s).
Using either method 1 or 2, you'll obtain the standard deviation (σ or s) necessary to calculate the
Cpk for your sheet metal bracket hole diameter.
