Study guide

Type B Distribution Models

ASQ Certified Calibration Technician CCT

Track your progress as you read

Finished sections get a check mark.

1 SI Definitions and Realizations

The Seven SI Base Units and the 2019 Revision Realizing SI Units in the Laboratory

2 Derived Units, Prefixes, and Constants

Derived Units and Physical Calculations SI Prefixes and Compound-Unit Conversions Physical Constants and Their Applications

3 Measurement Standards and Selection

Types and Roles of Measurement Standards Selecting Standards for an Operational Hierarchy

4 Traceability and Standard Substitution

Traceability Evidence and Impact Review Evaluating and Authorizing Standard Substitution

5 Dimensional and Mechanical Metrology

Linear Displacement and Dimensional Instruments Geometry, Thermal Expansion, and Mass Force, Torque, Pressure, Density, and Viscosity

6 Temperature, Humidity, Volume, and Flow

Temperature Sensors and Thermometer Corrections Humidity, Volume, and Flow

7 Electrical, Optical, and Time Measurements

Voltage, Current, Resistance, and AC Measurement Time, Frequency, Light, Power, and Energy

8 Measurement Methods

Direct, Indirect, and Ratio Methods Transfer, Differential, and UUT Substitution Methods

9 Instrument Characteristics and Measurement Data

Sensitivity, Variability, and Precision Bias, Linearity, Stability, and Hysteresis Data Formats, Readability, and Suitability Resolution, Confidentiality, and Data Integrity

10 Specifications and Error Sources

Interpreting IM&TE Specification Terms Specification Calculations and Operating Qualifiers Instrument, Operator, and Circuit Errors Environmental, Geometric, and Interference Errors

11 Measurement Assurance

Check Standards and Statistical Process Control Proficiency Comparisons and Gage R&R

12 Procedures and Governing Requirements

Procedure Structure and Published Practice Sources Method Development, Deviations, and Handling Regulations, Contracts, and Requirement Precedence

13 Instrument Adjustment and Standardization

Calibration, Zero, and Span Adjustment Linearization, Nulling, and Digital Adjustment

14 Calibration Environment

Temperature and Humidity Control ESD, Dust, Vibration, and Pressure

15 Operations, Roles, and Scheduling

Receiving, As-Found Testing, and Calibration Flow Asset Identification, Tracking, and Release Laboratory Roles and Initial Scheduling Interval Review, Overdue Control, and Workloads

16 Method and Software Validation

Method Verification and Validation Studies Software, Firmware, and Spreadsheet Validation

17 Records and Formal Reporting

Controlled Records and Calibration Certificates Status Labels and Out-of-Tolerance Reporting

18 Uncertainty Concepts and Decision Risk

Error, Uncertainty, and Measurand Definition TAR, TUR, Tolerance Consumption, and Capability Conformity Decisions and False-Accept Risk Guard Bands and Conformity Reporting

19 Uncertainty Models and Budgets

Measurement Models and Uncertainty Contributions Distribution Selection and Budget Assembly

20 Type A and Type B Evaluation

Type A Statistics and Repeatability Outlier Review and Worked Type A Analysis Type B Distribution Models

21 Uncertainty Reporting and Applied Mathematics

Propagation, Covariance, and Effective Degrees of Freedom Coverage Factors and Expanded-Uncertainty Reporting Notation, Unit Conversions, Ratios, and Decibels Rounding, Algebra, Graphs, and Angular Conversions

22 Quality Systems and Tools

Quality-System Components, Planning, and Improvement The Seven Basic Quality Tools

23 Audits and Corrective Action

Audit Types, Planning, and Nonconformance Control Impact Assessment, Root Cause, and Effectiveness

24 Safety, Standards, and Professional Conduct

Electrical, Pressure, Chemical, and Laboratory Hazards PPE, Safety Data Sheets, 6S, and Safe Restoration Quality Standards, GUM, and VIM Accreditation Scope and Professional Ethics

Type B Distribution Models

Key Takeaways Explain This Topic AI

Symmetric rectangular, triangular, and arcsine bounds give standard uncertainties a/√3, a/√6, and a/√2.

For the same half-width, the arcsine model gives greater standard uncertainty than the rectangular or triangular model.

Convert a reported expanded uncertainty to standard uncertainty using its stated coverage factor.

Last updated: October 2026

Many contributions, such as certificate uncertainty or specification bounds, are evaluated using Type B methods . The classification follows the evaluation method. Environmental variation, drift, or repeatability can instead be evaluated statistically when adequate observation data are available; a physical source is not inherently Type A or Type B.

According to GUM Clause 4.3, a Type B evaluation relies on the metrologist's scientific judgment using all available information, including:

  • Previous calibration data and historical control charts
  • Manufacturer technical specifications
  • Calibrated values and expanded uncertainties from accredited calibration certificates
  • Uncertainties assigned to reference data taken from physical handbooks
  • Digital display resolution and quantization limits

To combine a Type B uncertainty component with Type A components in a combined uncertainty budget, the non-statistical tolerance or limit ± a \pm a ± a must be converted into an equivalent standard uncertainty u ( x i ) u(x_i) u ( x i ​ ) , which represents one standard deviation ( σ \sigma σ ). This conversion requires selecting an appropriate mathematical model: the probability distribution .

Master Summary of Probability Distributions in Metrology

Distribution Shape Description Half-Width Symbol Divisor to Obtain u ( x i ) u(x_i) u ( x i ​ ) Formula for Standard Uncertainty u ( x i ) u(x_i) u ( x i ​ ) Typical Metrological Applications

Rectangular (Uniform) Flat plateau between − a -a − a and + a +a + a ; zero outside a a a 3 ≈ 1.73205 \sqrt{3} \approx 1.73205 3 ​ ≈ 1.73205 u = a 3 ≈ 0.577 a u = \frac{a}{\sqrt{3}} \approx 0.577 a u = 3 ​ a ​ ≈ 0.577 a Digital resolution ( ± 0.5 LSD \pm 0.5\text{ LSD} ± 0.5 LSD ), environmental chamber stability ( ± Δ T \pm \Delta T ± Δ T ), manufacturer specs without distribution or confidence level

Triangular Symmetrical triangle peaking at center, tapering to zero at ± a \pm a ± a a a a 6 ≈ 2.44949 \sqrt{6} \approx 2.44949 6 ​ ≈ 2.44949 u = a 6 ≈ 0.408 a u = \frac{a}{\sqrt{6}} \approx 0.408 a u = 6 ​ a ​ ≈ 0.408 a Two independent equal uniform contributions, or other evidence supporting a triangular model

U-Shaped (Arc-Sine) Bimodal bathtub curve peaking at ± a \pm a ± a , minimum at center a a a 2 ≈ 1.41421 \sqrt{2} \approx 1.41421 2 ​ ≈ 1.41421 u = a 2 ≈ 0.707 a u = \frac{a}{\sqrt{2}} \approx 0.707 a u = 2 ​ a ​ ≈ 0.707 a RF and microwave transmission line mismatch, sinusoidal power fluctuations, uniform random phase of a sinusoidal influence

Normal (Gaussian, 95%) Bell curve with stated 95 % 95\% 95% confidence interval a a a 1.960 ≈ 2.000 1.960 \approx 2.000 1.960 ≈ 2.000 u = a 1.96 ≈ a 2 u = \frac{a}{1.96} \approx \frac{a}{2} u = 1.96 a ​ ≈ 2 a ​ Manufacturer specifications quoting 95 % 95\% 95% confidence, expanded uncertainty from accredited certificates with k = 2 k=2 k = 2

Normal (Gaussian, 99.73%) Bell curve with stated 3 σ 3\sigma 3 σ ( 99.73 % 99.73\% 99.73% ) limits a a a 3.000 3.000 3.000 u = a 3 ≈ 0.333 a u = \frac{a}{3} \approx 0.333 a u = 3 a ​ ≈ 0.333 a Conservative manufacturer tolerances stated as 3 σ 3\sigma 3 σ or 99.73 % 99.73\% 99.73% limits

  • Detailed Mathematical Derivations and Applications
  • The Rectangular (Uniform) Distribution

The rectangular distribution is the default model under the GUM when a quantity is known to lie within upper and lower bounds [ − a , + a ] [-a, +a] [ − a , + a ] , but there is no specific knowledge about the probability of values within those bounds.

Mathematical Derivation of Variance

The probability density function (PDF) is constant across the interval: p ( x ) = { 1 2 a , − a ≤ x ≤ + a 0 , ∣ x ∣ > a p(x) = \begin{cases} \frac{1}{2a}, & -a \le x \le +a \\ 0, & |x| > a \end{cases} p ( x ) = { 2 a 1 ​ , 0 , ​ − a ≤ x ≤ + a ∣ x ∣ > a ​

Because the distribution is symmetric, the mean is zero ( E [ X ] = 0 E[X] = 0 E [ X ] = 0 ). The variance is the expected value of x 2 x^2 x 2 : V ( x ) = σ 2 = ∫ − a + a x 2 p ( x ) d x = ∫ − a + a x 2 ( 1 2 a ) d x = 1 2 a [ x 3 3 ] − a + a V(x) = \sigma^2 = \int_{-a}^{+a} x^2 p(x) dx = \int_{-a}^{+a} x^2 \left(\frac{1}{2a}\right) dx = \frac{1}{2a} \left[ \frac{x^3}{3} \right]_{-a}^{+a} V ( x ) = σ 2 = ∫ − a + a ​ x 2 p ( x ) d x = ∫ − a + a ​ x 2 ( 2 a 1 ​ ) d x = 2 a 1 ​ [ 3 x 3 ​ ] − a + a ​ V ( x ) = 1 2 a ( a 3 3 − ( − a ) 3 3 ) = 1 2 a ( 2 a 3 3 ) = a 2 3 V(x) = \frac{1}{2a} \left( \frac{a^3}{3} - \frac{(-a)^3}{3} \right) = \frac{1}{2a} \left( \frac{2a^3}{3} \right) = \frac{a^2}{3} V ( x ) = 2 a 1 ​ ( 3 a 3 ​ − 3 ( − a ) 3 ​ ) = 2 a 1 ​ ( 3 2 a 3 ​ ) = 3 a 2 ​

Taking the square root yields the standard uncertainty: u ( x ) = a 2 3 = a 3 ≈ a 1.73205 ≈ 0.5774 a u(x) = \sqrt{\frac{a^2}{3}} = \frac{a}{\sqrt{3}} \approx \frac{a}{1.73205} \approx 0.5774 a u ( x ) = 3 a 2 ​ ​ = 3 ​ a ​ ≈ 1.73205 a ​ ≈ 0.5774 a

Primary Calibration Applications

Digital Display Resolution (Quantization): A digital instrument rounds its internal estimate to the nearest display increment (least significant digit, LSD, or resolution d d d ). The maximum rounding error is half of the resolution: a = d / 2 = 0.5 LSD a = d/2 = 0.5\text{ LSD} a = d /2 = 0.5 LSD . u resolution = d / 2 3 = d 2 3 = d 12 ≈ 0.2887 d u_{\text{resolution}} = \frac{d / 2}{\sqrt{3}} = \frac{d}{2\sqrt{3}} = \frac{d}{\sqrt{12}} \approx 0.2887 d u resolution ​ = 3 ​ d /2 ​ = 2 3 ​ d ​ = 12 ​ d ​ ≈ 0.2887 d

Example: A digital indicator displays readings to 0.001 mm 0.001\text{ mm} 0.001 mm ( 1 μ m 1\ \mu\text{m} 1 μ m ). The semi-range is a = 0.0005 mm a = 0.0005\text{ mm} a = 0.0005 mm . The standard uncertainty due to resolution is: u res = 0.0005 mm 3 = 0.000289 mm = 0.289 μ m u_{\text{res}} = \frac{0.0005\text{ mm}}{\sqrt{3}} = 0.000289\text{ mm} = 0.289\ \mu\text{m} u res ​ = 3 ​ 0.0005 mm ​ = 0.000289 mm = 0.289 μ m

Manufacturer Specifications Without Stated Confidence: When an equipment manual quotes an accuracy of " ± 0.05 % \pm 0.05\% ± 0.05% of reading" with no mention of a normal distribution or confidence level, a rectangular model can be justified when reliable bounds are known and no better distribution information is available with half-width a = 0.05 % × Reading a = 0.05\% \times \text{Reading} a = 0.05% × Reading .

Environmental Temperature Stability: A laboratory thermostat maintains room temperature within ± 1.5 ∘ C \pm 1.5^\circ\text{C} ± 1. 5 ∘ C of setpoint. If no continuous logging histogram is available, assume a rectangular distribution with a = 1.5 ∘ C a = 1.5^\circ\text{C} a = 1. 5 ∘ C : u = 1.5 / 3 = 0.866 ∘ C u = 1.5 / \sqrt{3} = 0.866^\circ\text{C} u = 1.5/ 3 ​ = 0.86 6 ∘ C .

The Triangular Distribution

The triangular distribution applies when values are known to cluster near the center of the interval [ − a , + a ] [-a, +a] [ − a , + a ] and diminish linearly toward the extremes.

Mathematical Derivation of Variance

By convolution, the sum (sum or difference) of two independent rectangular distributions with identical half-widths a / 2 a/2 a /2 forms an exact symmetric triangle with base [ − a , + a ] [-a, +a] [ − a , + a ] .

The probability density function is: p ( x ) = { 1 a ( 1 − ∣ x ∣ a ) , − a ≤ x ≤ + a 0 , ∣ x ∣ > a p(x) = \begin{cases} \frac{1}{a} \left(1 - \frac{|x|}{a}\right), & -a \le x \le +a \\ 0, & |x| > a \end{cases} p ( x ) = { a 1 ​ ( 1 − a ∣ x ∣ ​ ) , 0 , ​ − a ≤ x ≤ + a ∣ x ∣ > a ​

Integrating x 2 p ( x ) x^2 p(x) x 2 p ( x ) across the interval yields the variance: V ( x ) = σ 2 = ∫ − a + a x 2 1 a ( 1 − ∣ x ∣ a ) d x = 2 ∫ 0 a ( x 2 a − x 3 a 2 ) d x = 2 [ a 3 3 a − a 4 4 a 2 ] = 2 ( a 2 3 − a 2 4 ) = a 2 6 V(x) = \sigma^2 = \int_{-a}^{+a} x^2 \frac{1}{a}\left(1 - \frac{|x|}{a}\right) dx = 2 \int_0^a \left(\frac{x^2}{a} - \frac{x^3}{a^2}\right) dx = 2 \left[ \frac{a^3}{3a} - \frac{a^4}{4a^2} \right] = 2 \left( \frac{a^2}{3} - \frac{a^2}{4} \right) = \frac{a^2}{6} V ( x ) = σ 2 = ∫ − a + a ​ x 2 a 1 ​ ( 1 − a ∣ x ∣ ​ ) d x = 2 ∫ 0 a ​ ( a x 2 ​ − a 2 x 3 ​ ) d x = 2 [ 3 a a 3 ​ − 4 a 2 a 4 ​ ] = 2 ( 3 a 2 ​ − 4 a 2 ​ ) = 6 a 2 ​

Taking the square root yields the standard uncertainty: u ( x ) = a 2 6 = a 6 ≈ a 2.44949 ≈ 0.4082 a u(x) = \sqrt{\frac{a^2}{6}} = \frac{a}{\sqrt{6}} \approx \frac{a}{2.44949} \approx 0.4082 a u ( x ) = 6 a 2 ​ ​ = 6 ​ a ​ ≈ 2.44949 a ​ ≈ 0.4082 a

A supported triangular model: Two independent, equal uniform components sum to an exact triangle. A backlash mechanism or controlled bath does not automatically have that shape; inspect the physics and data before selecting the divisor.

The U-Shaped (Arc-Sine) Distribution

The U-shaped distribution applies when a physical parameter varies cyclically or sinusoidally over time, dwelling primarily near the turnaround peaks and passing rapidly through the center.

Mathematical Derivation of Variance

Consider a variable undergoing simple harmonic oscillation with amplitude a a a : x ( t ) = a sin ⁡ ( ω t ) x(t) = a \sin(\omega t) x ( t ) = a sin ( ω t )

The probability of observing a value x x x in an infinitesimal window d x dx d x is inversely proportional to the speed of the wave: ∣ d x / d t ∣ = a ω cos ⁡ ( ω t ) = ω a 2 − x 2 |dx/dt| = a\omega \cos(\omega t) = \omega \sqrt{a^2 - x^2} ∣ d x / d t ∣ = aω cos ( ω t ) = ω a 2 − x 2 ​ . Normalizing over the domain ( − a , + a ) (-a, +a) ( − a , + a ) gives the arc-sine probability density function: p ( x ) = 1 π a 2 − x 2 , − a < x < + a p(x) = \frac{1}{\pi \sqrt{a^2 - x^2}}, \quad -a < x < +a p ( x ) = π a 2 − x 2 ​ 1 ​ , − a < x < + a

The variance is obtained by integrating: V ( x ) = σ 2 = ∫ − a + a x 2 π a 2 − x 2 d x V(x) = \sigma^2 = \int_{-a}^{+a} \frac{x^2}{\pi \sqrt{a^2 - x^2}} dx V ( x ) = σ 2 = ∫ − a + a ​ π a 2 − x 2 ​ x 2 ​ d x

Substituting x = a sin ⁡ θ x = a \sin\theta x = a sin θ (where d x = a cos ⁡ θ d θ dx = a \cos\theta d\theta d x = a cos θ d θ and a 2 − x 2 = a cos ⁡ θ \sqrt{a^2 - x^2} = a \cos\theta a 2 − x 2 ​ = a cos θ ): V ( x ) = 1 π ∫ − π / 2 + π / 2 ( a 2 sin ⁡ 2 θ ) d θ = a 2 π ∫ − π / 2 + π / 2 ( 1 − cos ⁡ 2 θ 2 ) d θ = a 2 π [ θ 2 ] − π / 2 + π / 2 = a 2 2 V(x) = \frac{1}{\pi} \int_{-\pi/2}^{+\pi/2} (a^2 \sin^2\theta) d\theta = \frac{a^2}{\pi} \int_{-\pi/2}^{+\pi/2} \left(\frac{1 - \cos 2\theta}{2}\right) d\theta = \frac{a^2}{\pi} \left[ \frac{\theta}{2} \right]_{-\pi/2}^{+\pi/2} = \frac{a^2}{2} V ( x ) = π 1 ​ ∫ − π /2 + π /2 ​ ( a 2 sin 2 θ ) d θ = π a 2 ​ ∫ − π /2 + π /2 ​ ( 2 1 − cos 2 θ ​ ) d θ = π a 2 ​ [ 2 θ ​ ] − π /2 + π /2 ​ = 2 a 2 ​

Taking the square root yields the standard uncertainty: u ( x ) = a 2 2 = a 2 ≈ a 1.41421 ≈ 0.7071 a u(x) = \sqrt{\frac{a^2}{2}} = \frac{a}{\sqrt{2}} \approx \frac{a}{1.41421} \approx 0.7071 a u ( x ) = 2 a 2 ​ ​ = 2 ​ a ​ ≈ 1.41421 a ​ ≈ 0.7071 a

Environmental cycling: A sinusoidal temperature variation sampled at a uniformly random phase has an arcsine distribution. On/off thermostat control alone does not prove this model; an exponential heating/cooling cycle can have another distribution.

Normal Distribution and Calibration Certificate Conversions

When a specification or calibration certificate quotes an uncertainty with a stated confidence interval, use the stated distribution and coverage information; a confidence statement alone does not prove a normal model.

Converting Expanded Uncertainty from a Calibration Certificate

Use the certificate’s stated expanded uncertainty, coverage factor, and coverage information. For a justified normal model with adequate degrees of freedom, k = 2 k=2 k = 2 corresponds to about 95.45% central coverage; it is not the interpretation of every certificate or distribution.

To enter this component into your working calibration uncertainty budget as a standard uncertainty u ( x i ) u(x_i) u ( x i ​ ) , simply divide by the stated coverage factor: u ( x i ) = U k u(x_i) = \frac{U}{k} u ( x i ​ ) = k U ​

Hypothetical exercise: A reference resistor certificate gives the following value, uncertainty, and coverage factor. These are invented teaching inputs rather than quotations from an actual NIST report: R std = 10 , 000.042 Ω , U = 0.024 Ω ( k = 2.00 , approx. 95 % confidence ) R_{\text{std}} = 10,000.042\ \Omega, \quad U = 0.024\ \Omega \quad (k = 2.00, \text{ approx. } 95\% \text{ confidence}) R std ​ = 10 , 000.042 Ω , U = 0.024 Ω ( k = 2.00 , approx. 95% confidence )

The standard uncertainty to enter into the laboratory's uncertainty budget is: u ( R std ) = 0.024 Ω 2.00 = 0.012 Ω u(R_{\text{std}}) = \frac{0.024\ \Omega}{2.00} = 0.012\ \Omega u ( R std ​ ) = 2.00 0.024 Ω ​ = 0.012 Ω

Converting Manufacturer Specifications with Stated Confidence Levels

If the stated interval is central coverage under a justified normal model, use the corresponding normal quantile. Confidence alone does not establish normality, infinite degrees of freedom, or an exact divisor. If a certificate supplies its coverage factor, use that factor and its stated interpretation. The following quantiles are normal-model examples:

90.00 % 90.00\% 90.00% Confidence: Divisor k = 1.645 ⟹ u = a / 1.645 k = 1.645 \implies u = a / 1.645 k = 1.645 ⟹ u = a /1.645

95.00 % 95.00\% 95.00% Confidence: Divisor k = 1.960 ⟹ u = a / 1.960 k = 1.960 \implies u = a / 1.960 k = 1.960 ⟹ u = a /1.960

95.45 % 95.45\% 95.45% Confidence ( 2 σ 2\sigma 2 σ ): Divisor k = 2.000 ⟹ u = a / 2.000 k = 2.000 \implies u = a / 2.000 k = 2.000 ⟹ u = a /2.000

99.00 % 99.00\% 99.00% Confidence: Divisor k = 2.576 ⟹ u = a / 2.576 k = 2.576 \implies u = a / 2.576 k = 2.576 ⟹ u = a /2.576

99.73 % 99.73\% 99.73% Confidence ( 3 σ 3\sigma 3 σ ): Divisor k = 3.000 ⟹ u = a / 3.000 k = 3.000 \implies u = a / 3.000 k = 3.000 ⟹ u = a /3.000

Comparative Worked Calculation: Impact of Divisor Selection

To illustrate how the choice of probability distribution affects calibration budgets, consider an instrument tolerance limit of ± a = ± 12.0 μ m \pm a = \pm 12.0\ \mu\text{m} ± a = ± 12.0 μ m . Let us calculate the resulting standard uncertainty under all four models:

Important

Critical Metrological Insight: For the exact same physical tolerance band ( ± 12.0 μ m \pm 12.0\ \mu\text{m} ± 12.0 μ m ):

The U-shaped distribution yields the largest standard uncertainty ( 8.49 μ m 8.49\ \mu\text{m} 8.49 μ m ) because the data concentrates near the extreme limits.

The triangular distribution yields a smaller standard uncertainty ( 4.90 μ m 4.90\ \mu\text{m} 4.90 μ m ) than the rectangular distribution because data concentrates near zero.

Mistakenly applying a triangular divisor ( 6 \sqrt{6} 6 ​ ) to a rectangular process underestimates the uncertainty by nearly 30 % 30\% 30% ( 4.90 4.90 4.90 vs 6.93 μ m 6.93\ \mu\text{m} 6.93 μ m ), potentially causing false-accept audit nonconformances!

Technical reference checked October 10, 2026: JCGM uncertainty and vocabulary publications .