Quality Control and Assurance Terminology: Definitions and Explanations

Accept / Reject Criteria

The predetermined standards used to determine whether a product, process, or service meets acceptable quality requirements. These criteria may include technical specifications such as dimensional tolerances, performance benchmarks, contractual obligations, or functional requirements. The concept assumes that measurable acceptance thresholds should be established for all processes and products, with regular assessments conducted to verify compliance with specifications.

Acceptable Quality Limit (AQL)

According to ISO 2859, AQL represents the worst tolerable quality level that can be considered satisfactory as a process average when evaluating a continuing series of production lots through acceptance sampling. It defines the maximum number of defective items permissible in a statistically selected sample before an entire batch is rejected. AQL is expressed as a percentage or ratio and serves as a benchmark to balance supplier quality performance with buyer risk tolerance.

Acceptance Number

The maximum quantity of nonconforming units or defects allowed in an inspection sample for the entire lot to be accepted. When the number of identified defects in the sample equals or falls below the acceptance number, the batch passes inspection. Exceeding this threshold typically indicates systematic quality problems requiring corrective action.

Acceptance Sampling Plan

A documented methodology that establishes the statistical approach for evaluating product quality, including sample size determination, inspection procedures, and acceptance or rejection criteria based on specified quality standards. Sampling plans may utilize single, double, multiple, sequential, or skip-lot sampling techniques, with different approaches applicable to attribute or variable data.

Adequacy of Standards

The suitability of a reference standard for calibrating measurement instruments and gauges used in quality control. Adequate standards should ideally be traceable to national or international measurement authorities such as the National Institute of Standards and Technology (NIST) to ensure accuracy and consistency across measurements.

Affinity Diagramming

A qualitative technique for organizing diverse ideas, observations, or data points into logical groupings based on natural relationships between items. Teams commonly use this method to identify patterns, themes, and commonalities within subjective information gathered from multiple sources or participants.

Agile Manufacturing

A production philosophy emphasizing flexibility and responsiveness to enable rapid product design, customization, and manufacturing. Agile manufacturing systems allow companies to deliver customized products at competitive prices and timelines without compromising quality or environmental standards, contrasting with traditional mass production approaches.

Analysis of Means (ANOM)

A statistical method applied in experimental design to identify process capabilities and problems by comparing treatment means against an overall average. ANOM helps determine whether specific process conditions or factors produce significantly different results in product characteristics.

Analysis of Variance (ANOVA)

A statistical technique that decomposes total variation in process outputs into components attributable to different sources or factors. ANOVA helps identify which variables significantly affect process performance, enabling targeted improvement efforts to reduce output variation and enhance consistency.

AOQ (Average Outgoing Quality)

The expected average quality level of products delivered to customers after inspection and acceptance sampling have been performed. AOQ calculation assumes that rejected lots undergo 100% inspection with defective items replaced or repaired, establishing the maximum percentage of defects that could reach customers under a given sampling plan.

Audit (Quality)

An objective, independent examination that compares actual quality performance, processes, or systems against established standards, specifications, or requirements. Quality audits verify compliance, identify gaps, and provide evidence-based findings to support improvement decisions.

Average Run Length (ARL)

The expected number of samples or subgroups that will be plotted on a control chart before detecting a process shift from its established control limits. ARL provides a statistical measure of how quickly a control chart will signal when a process moves out of statistical control.

Batch

A specific quantity of material or products manufactured under uniform conditions of production. Batches are typically identifiable through unique lot numbers or production codes for traceability purposes.

Bathtub Curve

A reliability curve depicting failure rates over a product’s lifecycle, characterized by three distinct phases: early high failure rates (infant mortality), a middle period of low constant failure rates (useful life), and increasing failure rates as products approach end-of-life (wear-out phase). The curve’s shape resembles a bathtub cross-section.

Bimodal Distribution

A frequency distribution displaying two distinct peaks or modes, indicating the presence of two different populations or operating conditions within the measured data. Bimodal distributions often suggest that data comes from mixed sources or that process conditions vary significantly.

Calibration

The process of comparing a measurement instrument against a known standard of verified accuracy to identify and document any deviations from required specifications. Calibration ensures measurement systems provide accurate, reliable data for quality control decisions.

Capability Process Index (Cp)

A statistical ratio comparing the specification tolerance range (upper specification limit minus lower specification limit) to the natural process variation (typically six standard deviations). Cp indicates whether a process has sufficient capability to consistently produce outputs within specification limits, with higher values indicating greater capability.

Causal Analysis

A systematic investigation methodology that identifies specific failures or defects, traces them to their root causes, and develops recommendations for eliminating defects by addressing underlying causes rather than symptoms.

Continuous Sampling Plan

An inspection approach designed for continuous production flows where inspection begins with 100% screening until a specified number of consecutive conforming units are found, then transitions to fractional sampling. When a nonconforming unit is detected during fractional sampling, the plan reverts to 100% inspection.

Control Chart

A graphical tool used in statistical process control to monitor process performance and capability over time by plotting measured data points against calculated control limits. Control charts help distinguish between common cause variation (inherent to the process) and special cause variation (indicating process changes).

Control Chart Factor

Mathematical constants derived from statistical probability distributions that facilitate the calculation of control limits for different types of control charts, such as X-bar and R charts.

Cost of Poor Quality (COPQ)

The total costs incurred from producing defective products or services, including both internal costs (scrap, rework, waste, inspection labor) and external costs (returns, warranty claims, recalls, repairs, liability, and brand damage). COPQ increases exponentially as defects move downstream through the supply chain, making early detection critical for cost minimization.

Cost of Quality (COQ)

The comprehensive costs associated with ensuring and maintaining quality, encompassing prevention costs, appraisal costs, and the costs of both internal and external failures. COQ provides financial justification for quality improvement investments by quantifying both the costs of conformance and nonconformance.

Critical Defect

A defect that, based on judgment and experience, poses safety hazards to users, creates potential for injury, violates regulatory requirements, or makes products legally unsaleable. Critical defects are typically assigned an AQL of zero, meaning any occurrence results in batch rejection.

Defect

Any nonconformance to customer requirements, specifications, or fitness for intended use. Defects are commonly classified by severity: critical (safety hazards), major (significant functionality or usability impact), and minor (cosmetic issues with minimal impact on performance or customer satisfaction).

Demerit Chart

A quality tracking tool that assigns weighted point values to different defect types (critical, major, minor) and calculates total demerit points for production lots. This approach provides a single metric for comparing quality levels across batches by accounting for both defect frequency and severity.

Design of Experiments (DOE)

A structured statistical methodology for planning, conducting, and analyzing controlled tests to understand how multiple factors affect process outputs. DOE enables efficient experimentation to identify optimal factor combinations that minimize variation and maximize desired outcomes. Key contributors include Genichi Taguchi and Ronald Fisher.

Double Sampling

An inspection technique where an initial sample (n1) is drawn and evaluated. If results are inconclusive (neither clearly acceptable nor rejectable), a second larger sample (n2) is drawn to reach a definitive acceptance or rejection decision.

Failure Mode and Effects Analysis (FMEA)

A proactive risk assessment methodology that systematically identifies potential failure modes in products or processes, evaluates their effects and causes, assesses the probability of occurrence, and prioritizes preventive actions. FMEA is widely used to design reliability into products and processes.

Inspection

The activities of measuring, testing, examining, or evaluating product or process characteristics and comparing results against specified requirements to determine conformance. Inspection verifies that outputs meet dimensional, functional, aesthetic, and performance specifications.

ISO 9000 Standards

A family of international quality management system standards developed by the International Organization for Standardization. ISO 9001 addresses requirements for quality management systems encompassing all organizational processes from design through service. ISO 9002 (now withdrawn) covered production and installation without design. ISO 9003 (now withdrawn) addressed only final inspection and testing.

Lot Tolerance Percent Defective (LTPD)

A quality level representing the maximum percentage of defects in a lot that a consumer is willing to accept with low probability. LTPD defines poor quality levels that should be rejected most of the time, complementing AQL which defines acceptable quality levels.

Lower Control Limit (LCL)

The boundary on a control chart positioned three standard deviations below the process average (centerline), representing the lower threshold for normal process variation. Data points falling below the LCL indicate special cause variation requiring investigation.

Minimum Acceptable Quality

The maximum defect rate or quality level that can be considered satisfactory as an average for process outputs. This concept establishes quality thresholds for acceptance sampling decisions.

Nondestructive Testing and Evaluation (NDE or NDT)

Inspection and testing techniques that assess product quality, material properties, or structural integrity without damaging or destroying the items being tested. Common NDE methods include ultrasonic testing, radiography, magnetic particle inspection, and visual inspection.

Number of Defective Units Chart (np Chart)

A type of attribute control chart used to monitor the total count of defective units from samples of constant size over time. The np chart helps identify process stability and detect shifts in defect rates.

Pareto Chart

A bar chart that displays frequency or impact data in descending order from left to right, often accompanied by a cumulative percentage line. Based on the Pareto Principle (80/20 rule), these charts help identify the vital few causes contributing most significantly to problems or effects.

Process Quality Audit

A systematic examination of process performance, controls, and operator practices compared against established standards and procedures. Process audits verify that processes operate as documented and that quality decisions follow defined criteria.

Q9000 Series

The American National Standards Institute (ANSI) and American Society for Quality (ASQ) version of ISO 9000 standards, designated as ANSI/ASQC Q9000-1 series. These standards provide U.S. equivalents for quality management systems, documentation, and audit procedures.

Quality Control

The operational activities and techniques employed to fulfill quality requirements, including monitoring processes, identifying defects, and taking corrective actions to ensure products or services meet specifications and customer expectations.

Quality Engineering

The application of statistical methods and engineering principles to design, develop, and maintain processes that operate efficiently, produce consistent outputs with minimal variation, and deliver defect-free products.

Random Sampling

A sampling method where items are selected such that every unit in the population has an equal probability of being chosen, and all possible sample combinations have equal likelihood of selection. Random sampling minimizes bias and supports valid statistical inference.

Reliability Engineering

The engineering discipline focused on predicting, preventing, and managing product or system failures throughout the lifecycle. Reliability engineering employs statistical analysis, testing, and design techniques to ensure products perform as intended under specified conditions over time.

Representative Sampling

A sampling approach that selects units in a manner that minimizes bias and ensures sample characteristics accurately reflect the entire population or lot from which they’re drawn.

Sample

A subset of units selected from a larger population or production lot for inspection or testing purposes. Samples are examined to make inferences about the quality of the entire population without inspecting every unit.

Statistical Process Control (SPC)

The application of statistical methods to monitor and control process performance over time. SPC techniques identify sources of variation, distinguish between common and special causes, and guide improvement actions to reduce variation and enhance process capability.

Statistical Quality Control (SQC)

A comprehensive term encompassing statistical techniques applied to quality management, including both statistical process control methods and broader quality measurement and improvement approaches.

Taguchi Methods

Quality engineering methodologies developed by Genichi Taguchi emphasizing robust design, loss functions, and experimental design. Taguchi methods focus on designing products and processes that perform consistently despite variations in operating conditions, based on the principle that any deviation from target specifications represents a loss to society.

Variation

The natural or induced differences observed between measurements, outputs, or process results. All processes exhibit some degree of variation, which quality control seeks to understand, minimize, and control to improve consistency and predictability.