CQT key terms

The total money spent on preventing poor quality, on appraising products against requirements, and on covering the losses caused by failures found internally or by customers.

A bar chart that ranks problem categories from largest to smallest frequency or cost, often with a cumulative line, so the few causes producing most of the effect stand out.

A branching picture that sorts possible causes of a problem into groups such as people, machine, method, material, measurement and environment, helping a team explore every plausible source before testing ideas.

A simple form printed in advance so data collectors tick or tally events in defined categories, making counts consistent and easy to summarize later.

A diagram of the steps in a process, drawn with standard symbols, used to understand how work actually flows and to spot redundancy, loops or gaps.

A graph plotting pairs of numerical values on two axes so the reader can judge whether two variables move together and roughly how strong that relationship is.

A bar chart of measurement data grouped into equal intervals, showing the shape, center and spread of a distribution and hinting at whether the process is stable.

A repeated questioning technique that asks why a problem occurred, then why that cause happened, and so on, until the underlying root cause is exposed.

A four-stage loop for improvement: plan a change, run it, check the results against the objective, then standardize or adjust and repeat the cycle.

An ongoing effort by everyone to make small incremental gains in processes, often called kaizen, rather than relying only on occasional large projects.

A time-ordered plot of a process measure with a center line and statistically calculated limits, used to separate routine variation from signals of an assignable cause.

The normal random ups and downs built into a process as currently designed; reducing it requires changing the system rather than reacting to individual data points.

Unusual variation from an identifiable source outside the routine process, such as a bad lot, a tired operator or a broken tool, showing up as a signal on a control chart.

A ratio comparing the width of the specification limits with the natural spread of a stable process, showing whether the process can meet tolerances; Cpk also accounts for centering.

A measure of spread equal to the square root of the average squared distance of data points from their mean, expressed in the same units as the data.

A symmetric bell-shaped curve in which most values cluster near the mean, with roughly 68 percent of values inside one standard deviation and about 95 percent inside two.

The arithmetic average, found by adding all values and dividing by how many there are; a measure of central tendency that is sensitive to extreme values.

Selecting units so that every item in the population has a known, non-zero chance of being chosen, which allows results to be extended to the whole lot.

Attribute data count or classify items (pass or fail, number of defects), while variable data are measured on a continuous scale such as length or weight; the two need different statistical tools.

A line graph of data in time order with a median line but no control limits, useful for spotting trends, shifts and runs before a full control chart is warranted.

Deciding whether to accept a lot by inspecting a sample drawn from it rather than every unit, using a plan that fixes sample size and accept or reject criteria.

A graph showing, for a given sampling plan, the probability of accepting a lot at each possible incoming quality level, revealing producer and consumer risk.

A full evaluation of the first units produced after a setup or change, checking every dimension and requirement to confirm the process is ready before volume output.

Inspection methods such as ultrasonic, radiographic, magnetic particle or dye penetrant testing that reveal internal or surface flaws without damaging the part being examined.

Any product, component or material that does not meet a specified requirement, which must be identified, segregated and formally dispositioned as scrap, rework, return or use-as-is.

How close a measurement result is to the true or accepted reference value; a measurement can be accurate only when systematic error, or bias, is small.

How closely repeated measurements of the same item agree with each other, reflecting random variation in the measurement system rather than its closeness to the true value.

Comparing an instrument's readings with a known reference standard of higher accuracy and recording the differences, so corrections or adjustments can be made.

An unbroken, documented chain of comparisons linking a measurement back to national or international standards, with each step carrying its own stated uncertainty.

A quantified range around a measured value that probably contains the true value, built from contributions of the instrument, the operator, the environment and the procedure.

The smallest change in a quantity that an instrument can display or distinguish; it sets a limit on what the measurement system can possibly detect.

A study that splits measurement variation into repeatability (same operator, same gage) and reproducibility (different operators), showing how much variation the gage adds.

A measuring device with known, certified characteristics, kept under controlled conditions and used to calibrate or verify the working instruments on the floor.

The set of documented processes, responsibilities and resources an organization uses to direct and control quality so that products consistently meet requirements.

A planned, independent examination of objective evidence to check whether activities and their results comply with planned arrangements and are actually effective.

Steps taken to remove the root cause of a detected nonconformity so the problem does not return, as distinct from simply fixing or reworking the affected item.

Action taken on a potential nonconformity, before it occurs, to remove the cause and lower the chance that the problem ever happens.

Procedures ensuring the right people hold the current approved version of each document, that changes are reviewed before release, and that obsolete copies are removed.

A top-level document describing an organization's quality management system, its scope, the processes it covers and how the applicable standard's requirements are addressed.