C-chart

Monitor the count of nonconformities in constant-size inspection units against Poisson-based center and control limits.

Version

v2 · 2026-09-19 · History

Domain-specific #

1424

Domain group

Formal Sciences

Origin domain

Experimental Design & Statistics

Subdomain

Attribute Control Charts → Experimental Design & Statistics

Aliases

C control chart, Count-of-nonconformities chart

Core Idea

A C-chart is an attribute control chart for a sequence of counts of nonconformities observed in inspection units of constant opportunity or size. A single unit can contain more than one nonconformity, so the observed variable is a defect count rather than the number or fraction of defective units. Under the conventional Poisson model, the in-control center is the mean count \(\bar c\), and three-sigma limits are commonly \(\bar c \pm 3\sqrt{\bar c}\), with a negative lower limit truncated to zero.

How would you explain it like I'm…

The Mistake-Counting Fence Chart

A toy factory checks one teddy bear each hour and counts every mistake on it, like a loose button or a crooked eye. They draw the counts on a chart with a middle line for the usual number and two fence lines above and below. If a count jumps outside the fences, workers go look for a special reason something went wrong.

Defect Count Watch Chart

A C-chart is a graph factories use to watch how many flaws they find each time they check something of the same size, like one roll of fabric. One item can have several flaws, so you count flaws, not broken items. You work out the average count from a time when things were running normally and draw it as a middle line. You also draw an upper and a lower line showing how far counts usually wander by chance. A point outside those lines is a signal to look for a special cause, but the lines don't tell you whether the product is good enough for customers.

Defect-Count Control Chart

A C-chart is a type of control chart used in quality control to track counts of nonconformities (defects) found in inspection units of the same size or opportunity. Because one unit can contain several defects, it counts defects rather than defective units, which is what an NP-chart does. Using data from a period when the process was stable, you compute the average count c-bar; under a Poisson model the variance equals the mean, so the limits are usually c-bar plus or minus three times the square root of c-bar, with a negative lower limit set to zero. A point beyond the limits or a defined pattern of points signals a possible assignable cause worth investigating. The limits describe expected natural variation, not engineering tolerances, and the chart checks stability rather than proving what caused a change. If units differ in size, you should switch to a U-chart, which uses counts per unit of opportunity.

 

A C-chart is an attribute control chart for a time-ordered sequence of nonconformity counts c_i from inspection units of constant opportunity or size. Since one unit can carry several nonconformities, the monitored variable is a defect count, not the number or fraction of defective units; that distinguishes it from an NP-chart, which counts nonconforming units in a fixed sample. Under the conventional Poisson model, the in-control center line is the mean count c̄, estimated from baseline data judged representative of stable common-cause operation, and three-sigma limits are c̄ ± 3√c̄, truncating a negative lower limit to zero. A point beyond a limit or a declared run-rule pattern prompts investigation of a possible assignable cause. The limits reflect expected process variation under the model, not engineering tolerance or customer acceptability. If exposure varies between intervals, the correct change is to normalize counts by opportunity and use a U-chart. The Poisson assumptions require comparable units and variance roughly equal to the mean; overdispersion, clustering, serial dependence, shifting defect definitions, rare very large opportunities, or baseline estimates contaminated by special causes can distort false-alarm rates. The chart monitors stability; it does not prove causation or guarantee conformance to specifications.

Scope of Application

The abstraction is literal wherever practitioners can identify the same constitutive roles, apply the same boundary tests, and obtain the same kind of output. The following habitats are uses of C-chart itself, not metaphors based only on resemblance.

• Manufacturing inspection. Tracking surface flaws or assembly nonconformities per constant item batch.
• Document quality. Monitoring error counts per constant-length or constant-opportunity record.
• Service processes. Counting defined incidents per comparable operating interval.
• Baseline establishment. Estimating an in-control center from screened historical observations.
• Ongoing surveillance. Detecting shifts after a process has entered routine operation.
• Chart selection. Choosing C only when both event type and exposure frame satisfy its count assumptions.

Clarity

A clear account of C-chart must preserve the recognition invariant stated in the Core Idea rather than rely on the title alone. Define one nonconformity and state whether several may occur in one unit. Demonstrate that inspection opportunities are effectively constant. Report the baseline interval, mean estimate, dispersion check, and signal rule. Keep control limits, specification limits, and response procedures conceptually separate. These declarations are not editorial extras: each changes what observations count, which transformations are licensed, and what conclusion can be drawn.

Manages Complexity

C-chart manages complexity by replacing a diffuse field of observations or possible operations with a bounded role structure: inspection unit supplies a constant-size area, item, interval, or opportunity frame makes counts comparable.; nonconformity definition supplies a stable operational rule determines what events enter the count.; count sequence supplies ordered observations preserve time and expose shifts or unusual points.; baseline mean supplies a stable reference period estimates the in-control expected count \(\bar c\).; poisson variance model supplies the expected variance supplies the square-root scale for limits..

Abstract Reasoning

1. Specify the unit and count rule before reviewing outcome values.
2. Check that exposure and detection opportunity remain comparable across observations.
3. Screen the baseline for obvious special causes without erasing inconvenient variation silently.
4. Estimate \(\bar c\) and evaluate whether Poisson dispersion is plausible.
5. Compute and label center and control limits, truncating only the impossible negative bound.
6. Apply the declared point and run rules to the ordered sequence.

Knowledge Transfer

The strict upward abstraction is Quality Control. C-chart instantiates Quality Control because it repeatedly compares observed process counts with a stable reference model to detect departures requiring investigation. Within attribute control charts, the full mechanism transfers literally when the same roles and boundary tests recur. Beyond that domain, only the parent-level skeleton should travel. Reusing the label C-chart after removing its constitutive vocabulary would hide a change of mechanism behind an analogy. The honest transfer rule is therefore two-stage: recognize the domain-specific pattern first, then lift only the parent relation that remains invariant under a substrate change.

Relationships to Other Abstractions

Current abstraction C-chart Domain-specific

Parents (1) — outgoing typed relations

• C-chart is a kind of Quality Control Prime

  C-chart instantiates Quality Control because it repeatedly compares observed process counts with a stable reference model to detect departures requiring investigation.

Hierarchy paths (2) — routes to 2 parentless roots

• C-chart → is a kind of Quality Control → is a kind of Verification → is a kind of Evaluation → presupposes Comparison

Show alternative path (1)

Neighborhood in Abstraction Space

C-chart sits in a sparse region of the domain-specific corpus (95^th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.

Family — Statistical Process Control Charts (7 abstractions)

Nearest neighbors

• Np-chart — 0.80
• Natural Process Variation — 0.80
• P-chart — 0.78
• Effective Data Transfer Rate — 0.77
• U-chart — 0.76

Computed from structural-signature embeddings · 2026-10-25