CHAPTER 5

UNCERTAINTY ANALYSIS

1
 Introduction

Measurement
Harmonization
Errors

Correction for
Design-Stage
Correlated
Uncertainty
Systematic Errors
Analysis

Multiple-
CONTENTS Identifying
Measurement Error
Uncertainty Analysis Sources

Higher-Order
Uncertainty Systematic
and Random
• Nth-Order Errors
Uncertainty

Uncertainty
Advanced-
Analysis:
Stage
Error
Uncertainty
Propagation
Analysis
2
Two accepted professional documents on uncertainty analysis

1. The American National Standards Institute/American

Society of Mechanical Engineers (ANSI/ASME) Power
Test Codes (PTC) 19.1 Test Uncertainty is the United States
engineering test standard.
2. The International Organization on Standardization’s ‘‘Guide
to the Expression of Uncertainty in Measurement’’ (ISO GUM)
is an international metrology standard.
• The two differ in:
- PTC 19.1 refers to random and systematic errors,
terms that classify errors by how they manifest themselves in
the measurement.
- ISO GUM refers to type A and type B errors, terms
that classify errors by how their uncertainties are estimated.
3
5.2. Measurement errors
The true value & the
measured data set,
containing both systematic
and random errors.

The systematic errors shift

the sample mean away from
the true mean by a fixed
amount

The random errors bring

about a distribution of
measured values about the
sample mean. 4
5.2. Measurement errors

x' = x  u x ( P%)
Uncertainty analysis ux

5
uncertainty analysis

𝑢𝑥 contains the uncertainties assigned to all known errors

1. The test objectives are known and the measurement itself is

a clearly defined process.
2. Any known corrections for systematic error have been
applied to the data set, in which case the systematic
uncertainty assigned is the uncertainty of the correction.
3. Assume a normal distribution of errors and reporting of
uncertainties.
4. The errors are assumed to be independent of each other.
5. The engineer has some ‘‘experience’’ with the system
components.
6
 Measurement errors

Errors can enter into a

measurement with emphasis on
their influences on the test
result.

Both random errors &

systematic errors are
considered.
7
 5.3 DESIGN-STAGE UNCERTAINTY ANALYSIS

Uncertainty analysis at this time is

used to assist in selecting equipment
and test procedures based on their
relative performance.

A measurement system consists of

sensors and instruments, each with
their respective contributions to system
uncertainty.
8
5.3 DESIGN-STAGE UNCERTAINTY ANALYSIS

Zero-order uncertainty of Instrument, u0:

Even all errors are zero, a measured value
must be affected by our ability to resolve the
information provided by the instrument.

u0: Amount due to instrument resolution; all

other aspects of the measurement are
perfectly controlled.

u0: Random uncertainty caused by the data

scatter due to instrument resolution

9
 5.3 DESIGN-STAGE UNCERTAINTY ANALYSIS

Assign a numerical value to u0 of one-half of the

analog instrument resolution or to equal to its digital
least count (reading with a probability of 95%).

Least Significant Digit of the readout

Error has a normal distribution with its uncertainty

applied equally to either side of the reading

10
uncertainty of Instrument, uc

Instrument uncertainty (uc) -

estimate of the expected
systematic uncertainty due to the
instrument.

If no probability level is
provided with such
information, a 95% level
can be assumed.
11
 5.3 DESIGN-STAGE UNCERTAINTY ANALYSIS

Combining Elemental Errors: Root-sum-squares RSS Method

• Uncertainty A measurement of x:
propagation: k elements of error, each of
uncertainty uk; k =1; 2; . . . K
Individual
measurement
A realistic estimate of the
error interacts uncertainty ux
with other
errors to affect
the uncertainty

12
 Combining Elemental Errors:
Design state uncertainty analysis, 𝑢𝑑

• An interval found by combining the

instrument uncertainty with the zero-order
uncertainty,

13
 5.3 DESIGN-STAGE UNCERTAINTY ANALYSIS

Example 5.1
• Consider the force measuring • An estimate of the instrument
instrument described by the uncertainty depends on the
following catalog data. uncertainty assigned to each of the
• Provide an estimate of the contributing elemental errors of
uncertainty attributable to this linearity,e1, hysteresis, e2,
instrument and the instrument
design-stage uncertainty.
With K= 2,
Resolution: 0.25 N
Range: 0 to 100 N
Linearity error: within 0.20 N over range instrument resolution = 0.25 N, assume
Hysteresis error: within 0.30 N over range u0 = 125 N.
• ASSUMPTIONS Instrument
uncertainty at 95% level; normal
distribution
• FIND uc, ud
14
 Example 5.2
Avoltmeter is used to measure the electrical output signal
from a pressure transducer. The nominal pressure is
expected to be about 3 psi (= 0.2 bar). Estimate the design-
stage uncertainty in this combination. The following
information is available:

15
• The uncertainty in the voltmeter • The uncertainty in the pressure
at the design stage transducer
• Consider the uncertainty in
each of the instrument
elemental errors of linearity, 𝑒1
• For a nominal pressure of 3 psi, sensitivity, 𝑒2
expect to measure output: 3 V

• Since
• Design-stage uncertainty in the
• 𝑢𝑑 of combined system (two
voltmeter
devices) using RSS method

• Since the sensitivity is 1 V/psi,

the uncertainty in pressure 16
 5.4 IDENTIFYING ERROR SOURCES

calibration
errors

Three error
data-
source acquisition
groups errors

data-reduction
errors

17
 Calibration Errors

• INCLUDING: 1. Standard/
reference value used in the
calibration,
• 2. Instrument/system under
calibration,
• 3. Calibration process.

• Laboratory standard used for

calibration contains some
• There can be a difference between
inherent uncertainty, and this
the value supplied by the
is passed along with the input standard and the value actually
value on which the calibration sensed by the measuring system.
is based.
• Measuring system errors:
linearity, repeatability,
hysteresis => contribute to
uncertainty.
18
Data-Acquisition Errors

19
Data-Reduction Errors

20
5.5 SYSTEMATIC AND RANDOM ERRORS

• differ on repeated
Random measurements
• lead to data scatter
errors • be quantified by statistical
methods

• remain constant in repeated

Systematic measurements
• under fixed operating
errors conditions

21
 Random Error

• Random standard uncertainty, 𝑠𝑥ҧ defined by the

interval ±𝑠𝑥ҧ

with degrees of freedom  = N - 1

• Random uncertainty at a desired confidence
level is defined by the interval

22
 SYSTEMATIC ERRORS

Systematic standard uncertainty: Estimate of the

range

Its interval has a confidence level of one standard

deviation, equivalent to a probability level of 68% for a
normal distribution.

Confidence level is given by tn; Pb, or simply t. The

interval defined by the systematic uncertainty

95% probability level

23
5.6 UNCERTAINTY
Errors are
ANALYSIS: ERROR quantified by
PROPAGATION uncertainties

24
uncertainty analysis error propagation

25
5.7 ADVANCED-STAGE UNCERTAINTY ANALYSIS

26
Endues of uncertainty analysis in
*design of measurement systems, through selection of equipment
* procedure, in the interpretation of measured data.

Presenting procedures for estimating

uncertainty at each of the stages.

The procedures will provide reasonable

estimates of the uncertainty to be
expected in measurements.
27
 • estimate uncertainty
Procedures of propagation within a
uncertainty
analysis measurement

Propagation of • determine a result

uncertainty through some
among independent functional relationship
measured variables

28
 DESIGN-STAGE UNCERTAINTY ANALYSIS

(1) Design stage *(2) Advanced

*(3) Multiple
stage and/or single
measurement.
measurement

refers to the initial analysis performed

prior to the measurement

29
30
 Example 5.8
Ten repeated measurements of force F are made over time
under fixed operating conditions.
Estimate the random standard uncertainty due to the
elemental error in estimating the true mean value of the
force based on the limited data set.

31
Example 5.8

• Estimate the random

standard uncertainty
due to the elemental error in
estimating the true mean value of the
force based on the limited data set.

• Estimate the
systematic
uncertainty in the
measurement
instrument.

32
 Example 5.8
• The mean value of • Error enters the
the force based on measurement during
finite data data acquisition
• Random standard
uncertainty in this
• A random error is elemental error is
associated with computed through the
assigning the sample standard deviation of
mean value as the the means
true value of force
because of the small
data set
33
 Example 5.1
• Consider the force • Resolution: 0.25 N
measuring instrument
described by the following
• Range: 0 to 100 N
catalog data. • Linearity error:
• Provide an estimate within 0.20 N over
of the uncertainty range
attributable to this • Hysteresis error:
instrument and the
within 0.30 N over
instrument design-stage
uncertainty. range

34
From Example 5.1

• Lacking information?
• Systematic standard
uncertainties?

• b = ????
• tv,95 = ?????
• B = ????

35
Propagation of Uncertainty to a Result

• True value:
• Means value:

• Uncertanties
• Propagation of random Uncertainty

36
• EXERCISES:
5.54,
5.46,
5.30,
5.26,
5.19,
5.17,
5.10

37