IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, VOL. 61, NO. 8, AUGUST 2012 2107
Measurement Fundamentals: A Pragmatic V
Luca Mari, Paolo Carbone, Senior Member, IEEE, and Dario Petri, Fellow, IEEE
Abstract—Measurements are more and more required in an
increasing variety of human activities to acquire reliable informa-
tion useful for effectively supporting decision-making processes.
Furthermore, the entities whose properties are to be measured
and the measuring systems are becoming increasingly complex,
hardly to be modeled and managed. In thisrapidly evolving
scenario, several issues concerning the fundamentals of measure-
mentscience and technology arise. The purpose of this paper
is to discuss some aspects of the crucial question about which
evaluation processes can be considered measurements. Rather
than focusing on formal conditions or technological constraints,
we propose a pragmatic characterization of measurement, under
the assumption that a better comprehension of the concept can
be achieved by identifying and discussing the basic features which
justify the reliability attributed to measurement results. With such
an interdisciplinary approach, this work aims at promoting a
broad discussion among interested researchers, even working in
different scientific disciplines, so to increase the synergies among
different research areas and to improve the body of knowledge
about measurement fundamentals.
Index Terms—M/Measurement, M/Modeling, M/Multivariable
systems, U/Uncertain systems, U/Uncertainty.
I. INTRODUCTION
I
N MANY, if not all, human activities, measurement is con-
sidered a fundamental process to obtain reliable information
on the empirical world,in view of the Galilean motto of
measuring what is measurable and making measurable what is
notyet.Nowadays, a systematic adoption of measurement is
particularly solicited by the widespread application of techno-
science in the social, industrial, economical, . . .fields.Once
specifically aimed at evaluating physical quantities, measure-
ments are, today, more and more required in biology, medicine,
economy, sociology, psychology, . . .Furthermore, measuring
systems are becoming complex entities, in which the customary
measurement techniques have to be extended to multivariate
measurands, and the measurement of physical quantities has
to be often complemented by the measurement of nonphysical
properties, sometimes called “weakly defined measurement”
[1] or simply “soft measurement.” An excellent reference on
this matter is the document Evolving Needs for Metrology in
Manuscript received November 9, 2011; revised January 4, 2012; accepted
January 9, 2012.Date of publication April 30, 2012; date of current version
July 13,2012.The Associate Editor coordinating the review process for this
paper was Dr. Wendy Van Moer.
L. Mari is with Università Carlo Cattaneo (LIUC), 21053 Castellanza, Italy
(e-mail: lmari@liuc.it).
P. Carbone is with the University of Perugia, 06125 Perugia, Italy (e-mail:
carbone@diei.unipg.it).
D. Petriis with the University of Trento,38050 Trento, Italy (e-mail:
petri@disi.unitn.it).
Digital Object Identifier 10.1109/TIM.2012.2193693
Trade, Industry and Society and the Role of the BIPM [2],
states, for example, that,currently, “an estimated 80% (of the
world trade) is affected by standards and regulations” and
according to various studies, “the cost to producers and se
providers of complying with standards can be 10% of prod
tion costs.” Of course, measurement is the basis to assess
compliance. The Bureau International des Poids et Mesures
(BIPM) document lists some of the application areas where
the role of measurement is increasingly critical: They inclu
“transport; information technology, navigation, and telecom-
munications; electronics and optics; electromagnetic and i
izing radiation; energy; climate change and environmental
pollution control; clinical chemistry and laboratory medicin
food safety; antidoping; pharmaceuticals; and forensics and
security.”
This evolving scenario arises several significant issues for
measurement science, not only at the operative level—for ex-
ample, about the possibility to apply the now-standard proce
dures of uncertainty evaluation specified by the Guide to th
Expression of Uncertainty in Measurement [3] in such dive
fields—but also, and primarily, in reference to the state an
nature of measurement science itself. A fundamental critical
problem relates to the very concept of measurement: Is its
acceptation, as commonly understood in the measurement of
mechanical, optical, thermal, electrical, . . . quantities, already
adequate and directly applicable in this broader context? T
question is not only lexical, i.e.,whether a single entry in a
vocabulary may accommodate all usages of the term or only
related to the, however important, goal of mutual understand-
ing,which is the basic goal thatdrove a number of leading
international organizations to gather into the Joint Commit
for Guidesin Metrology and to produce the International
Vocabulary of Metrology (VIM3) [4]. Rather, an inquiry on a
widely shared meaning of “measurement” is useful to give a
convincing justification to the customary claim of the “spec
reliability” of measurement itself, which is surely not assum
in the case of, e.g.,subjective judgment or guess: The public
trust attributed to measurement results and the resource s
ing acceptable for measurement should not be approved for
such other activities.
If the generic process of assigning a quantity value to a
quantity of interest is termed “evaluation” (as in the abstract
case of “function evaluation”), so thatmeasurement, subjec-
tive judgment, and guess are all examples of evaluations, the
problem may be then formulated: Whatdoes itcharacterize
measurement as a specific kind of evaluation?
Since the problem is not conceptually new (although per
this formulation is), some answers have been already prop
in the past. At least three well-known general standpoints
be mentioned.
0018-9456/$31.00 © 2012 IEEE