What is a P-Chart? Definition & Manufacturing Examples

What is a P-Chart?

A p-chart (proportion chart) is an attribute data control chart used in statistical process control to monitor the proportion of defective items in a sample over time. Each item in the sample is classified as either defective or non-defective, and the fraction defective (p) is plotted for each sample period. The p-chart is the most versatile of the attribute control charts because it handles both constant and variable sample sizes.

The p-chart is based on the binomial distribution, which describes the probability of observing a specific number of defective items in a sample of n items when each item has the same probability p of being defective. The control limits are derived from this distribution and set at three standard deviations from the overall average proportion defective (p-bar).

When sample sizes vary — which is common in manufacturing where production volumes fluctuate daily — the control limits change with each sample because the standard deviation of a proportion depends on sample size. Larger samples produce tighter control limits because there is more statistical certainty. This variable-limit feature is what distinguishes the p-chart from the np-chart, which requires constant sample sizes.

How a P-Chart Works in Manufacturing

Creating a p-chart begins with defining what constitutes a defective item. Clear, specific criteria are essential — every inspector must classify items the same way. Ambiguous criteria produce unreliable data.

For each sample period (shift, hour, lot), inspect n items and count the number of defective items (d). Calculate the proportion defective: p = d / n. Collect at least 20 to 25 samples to establish the baseline.

Calculate the overall proportion defective: p-bar = total defective / total inspected.

For each sample, calculate the control limits:

• UCL = p-bar + 3√(p-bar × (1 - p-bar) / n)
• LCL = p-bar - 3√(p-bar × (1 - p-bar) / n) or 0 if negative

Because the limits depend on n, samples with different sizes have different limits. Many manufacturers simplify this by using the average sample size for limit calculations, then only recalculating when a specific sample size differs from the average by more than 25%.

Plot each sample's proportion defective against the limits. Points above the UCL indicate process deterioration. Points below the LCL indicate possible process improvement worth investigating and sustaining.

P-Chart Example

A metal stamping operation runs three shifts with different production volumes. The quality team inspects all units produced each shift and classifies each as acceptable or defective.

Overall: p-bar = 99 / 2,310 = 0.0429

For the Night shift with n = 320:

• UCL = 0.0429 + 3√(0.0429 × 0.9571 / 320) = 0.0429 + 0.0339 = 0.0768

The Night shift readings of 0.069 and 0.081 show that the second Night shift exceeds the UCL. Investigation reveals that the Night shift operates with reduced lighting and fewer experienced operators, leading to higher defect rates. The corrective action includes improved lighting and additional operator training.

Why P-Charts Matter for Production Scheduling

P-charts provide schedulers with real-time quality intelligence that affects capacity planning. When the proportion defective increases, more units must be produced to meet net order quantities, consuming additional machine time and materials.

The p-chart's ability to handle variable sample sizes makes it particularly useful for job shop environments where lot sizes vary from order to order. Scheduling software like Resource Manager DB can incorporate yield rates based on p-chart data to calculate realistic production quantities.

P-chart data that reveals shift-to-shift or machine-to-machine differences in defect rates helps schedulers make informed routing decisions — assigning quality-critical jobs to shifts or machines with the best quality performance.

Related Terms

• NP-Chart — tracks number defective with constant sample sizes instead of proportions
• C-Chart — tracks defect counts per unit rather than defective units
• Attribute Data — the data type used in p-charts and other attribute control charts

FAQ

---

This term is part of our Manufacturing & Production Scheduling Glossary. Learn more about quality control, scheduling, and manufacturing terminology.