Vous êtes sur la page 1sur 5

A Portrait of an ISO STEP

Allison Barnard Feeney


Systems Integration Division,
Engineering Laboratory,
Tolerancing Standard
National Institute of Standards and Technology,
Gaithersburg, MD 20899 as an Enabler of Smart
e-mail: allison.barnardfeeney@nist.gov

Simon P. Frechette
Manufacturing Systems
Systems Integration Division,
Engineering Laboratory,
The International Organization for Standardization (ISO) has just completed a major
National Institute of Standards and Technology,
effort on a new standard ISO 10303-242 titled “Managed Model Based 3D Engineering.”
Gaithersburg, MD 20899
It belongs to a family of standards called STEP (STandard for the Exchange of Product
e-mail: simon.frechette@nist.gov
model data). ISO 10303-242 is also called the STEP Application Protocol 242 (STEP AP
242, for short). The intent of STEP AP 242 is to support a manufacturing enterprise with
a range of standardized information models that flow through a long and wide “digital
Vijay Srinivasan thread” that makes the manufacturing systems in the enterprise smart. One such standar-
Fellow ASME
dized information model is that of tolerances specified on a product’s geometry so that
Systems Integration Division,
the product can be manufactured according to the specifications. This paper describes
Engineering Laboratory,
the attributes of smart manufacturing systems, the capabilities of STEP AP 242 in han-
National Institute of Standards and Technology,
dling tolerance information associated with product geometry, and how these capabilities
Gaithersburg, MD 20899
enable the manufacturing systems to be smart. [DOI: 10.1115/1.4029050]
e-mail: vijay.srinivasan@nist.gov
Keywords: ISO, standards, STEP AP242, 3D models, PMI, tolerances, smart manufacturing,
presentation, representation

1 Introduction It is important to observe that ISO 16792 is mainly a presenta-


tion standard. A human being, trained in the field of engineering
Progress in the field of computer aided tolerancing is strongly
product documentation, can read Fig. 1 and interpret it for further
tied to the evolution of national and international dimensioning
action. In fact, a model such as the one shown in Fig. 1 can be
and tolerancing standards [1]. These standards lay down the rules
zoomed, panned, and rotated to view any detail. But the trend, as
for the specification of geometric information necessary to manu-
we will see in Sec. 2, is to make such information machine-
facture parts so that they can be assembled into properly function-
readable. To accomplish this, there is an urgent need to have a
ing products. The task of developing such international standards
computable representation of the type of information in Fig. 1 so
has been shared by several technical committees (TC)—and their
subcommittees (SC)—of the ISO. The works of three such
committees are of interest to this paper.
The international standards for tolerancing symbols and their
definitions (essentially, the syntax and semantics of dimensional
and geometric tolerancing) are developed by ISO TC 213
“Dimensional and Geometrical Product Specifications (GPS) and
Verification.” Its standards are sometimes referred to as ISO GPS
standards, which also include verification standards that deal with
dimensional and coordinate metrology. The current status and
future plans for ISO GPS standards were presented in the 12th
CIRP Computer Aided Tolerancing Conference [2] and so will
not be repeated here.
Since three-dimensional (3D) models are rapidly replacing two-
dimensional (2D) drawings as the master for product technical
data in manufacturing industry [3], a need arose for standardized
indications of dimensions and tolerances on 3D models. This need
was met by ISO TC 10 “Technical Product Documentation” in its
standard ISO 16792 [4]. Figure 1 illustrates an example of how to
present the dimensioning, tolerancing, surface texture, and other
similar information as “product manufacturing information
(PMI)” on a 3D model according to ISO 16792.

Contributed by the Computers and Information Division of ASME for publication


in the JOURNAL OF COMPUTING AND INFORMATION SCIENCE IN ENGINEERING. Manuscript
received September 30, 2014; final manuscript received October 27, 2014; published
online April 8, 2015. Editor: Bahram Ravani.
This material is declared a work of the U.S. Government and is not subject to
copyright protection in the United States. Approved for public release; distribution is Fig. 1 An example of standardized presentation of PMI on a 3D
unlimited. model [4]

Journal of Computing and Information Science in Engineering JUNE 2015, Vol. 15 / 021001-1

Downloaded From: http://computingengineering.asmedigitalcollection.asme.org/ on 01/13/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use


that an engineering information system (such as computer aided which only “model-based enterprise” is highlighted immediately
manufacturing) can process it automatically to take further action. below that node as an enabler because that is the one relevant to
The task of developing a standardized computable representation, this paper. Similarly, model-based enterprise has many enablers
in the form of standardized data models, rests with ISO TC 184/ including “3D model-based engineering,” which has many ena-
SC 4 Industrial Data. blers including “ISO STEP AP 242 standard,” which contains
ISO TC 184/SC 4 develops a family of standards, the most many components including “PMI representation standard.” We
renowned of which is called STEP that describes standardized data will explain each of the nodes highlighted in Fig. 2 in some detail.
models in several APs (Application Protocols). It has just com- The enablers highlighted in Fig. 2 have been selected after con-
pleted a major development of STEP AP 242 for “Managed Model siderable discussion and debate with stakeholders. In each of the
Based 3D Engineering.” STEP AP 242 contains computable repre- past 5 years, the National Institute of Standards and Technology
sentations for several types of 3D model data, including dimen- (NIST) has been hosting a summit of experts from government,
sional and geometric tolerances. Thus we are moving from mere industry, and academia on “model-based enterprise” [7]. A com-
presentation of PMI in 3D models to representation of such infor- mon theme that has run through these annual summit meetings is
mation that is machine-readable. And this is considered a major that the world is moving (1) away from paper documentation and
breakthrough that will enable several innovations in manufacturing. drawings and (2) toward information models that are machine-
This paper describes how STEP AP 242 PMI representation readable. In short, what we envision as a smart manufacturing sys-
takes an important step toward making future manufacturing sys- tem is enabled, among other things, by a model-based enterprise
tems “smart.” Section 2 gives a brief description of smart manu- that uses computable models both within a factory and across a
facturing systems and some of their enablers. This is followed in supply network.
Sec. 3 by an introduction to the new ISO STEP AP 242. Section 4 An important enabler of a model-based enterprise is the 3D
then addresses PMI standardization covered by AP 242 and makes model-based engineering, in which a digital 3D model serves as
the case for its portrayal as an enabler of smart manufacturing sys- the authoritative information source for all activities in the prod-
tems. Some of the future plans and challenges are discussed in uct lifecycle and across the supply network (it is important to
Sec. 5 and a summary is provided in Sec. 6. point out that there are also several other important enablers of
model-based enterprise, such as model-based systems engineer-
ing, but these are not covered in this paper). The advantages of 3D
2 Smart Manufacturing Systems model-based engineering, and the associated R&D challenges, are
We live in a new manufacturing era that has been called the discussed and updated every year during the model-based enter-
“Third Industrial Revolution [5],” which is characterized by the prise summit at NIST. Many of the major findings of these summit
digitization of manufacturing. In this era of digitization of manu- meetings have remained relatively stable. For example, the con-
facturing, information will play an important role. How the prod- clusions from the fourth NIST summit held in December 2012 are
uct’s design and manufacturing information is authored, [7]
exchanged, and processed will determine who will succeed in this • Model-based methods and tools are increasing manufacturing
era. The metaphor “digital thread” has been recently invoked to productivity, but challenges remain in composite structure
picture the flow of information along the product lifecycle and manufacturing, additive manufacturing, and process data
across the supply network [6]. sharing.
Manufacturing systems in this new era will have to get “smart.” • Model-based inspection combined with optical measurement
They need to be autonomous, self-aware, and self-correcting. In techniques is a potential game changer.
short, they should be able to function with as little human inter- • Light-weight visualization formats are making model-based
vention as possible, while at the same time work harmoniously engineering feasible for small and medium sized companies.
with human supervision and collaboration. Reaching this goal will • Systems engineering is an increasingly important component
be a daunting task, but research and development (R&D) is under- of model-based enterprise.
way in several areas to pave the path toward this goal. Some of • Open standards and reference implementations are critical.
the R&D efforts relevant to this paper can be explained using a
hierarchy of technical enablers shown Fig. 2, in which only a few The last conclusion provides a strong justification for the pres-
nodes along a path in the hierarchy are highlighted. A “smart ence of ISO STEP AP 242 standard in Fig. 2 as an enabler of 3D
manufacturing system” in this diagram has many enablers, out of model-based engineering, and a good segue to Sec. 3.

3 ISO STEP AP 242


The most commonly implemented and used APs of the ISO
STEP standards are AP 203 [8] and AP 214 [9], whose latest edi-
tions were published in 2011 and 2010, respectively. Both these
APs address the same standardized 3D product models. It is gen-
erally perceived, however, that AP 203 was driven by the aero-
space and defence industry, and AP 214 was driven by the
(German) automotive industry. In 2009, a decision was made to
merge these two APs into one called AP 242. This is the genesis
of ISO 10303-242 “Managed Model Based 3D Engineering” [10],
which is currently in the final stages of the ISO administrative
process for publication as an International Standard.
Merging two popular APs into one has several advantages,
including
• Optimizing standards development and maintenance resour-
ces by eliminating duplication of efforts.
• Introducing new functionalities that are common to many
industry sectors.
• Strengthening the acceptance and support by manufacturing
Fig. 2 A partial, hierarchical enabler diagram industry by establishing a single universal brand.

021001-2 / Vol. 15, JUNE 2015 Transactions of the ASME

Downloaded From: http://computingengineering.asmedigitalcollection.asme.org/ on 01/13/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use


Figure 3 illustrates the contents of ISO STEP AP 242.
A quick study of Fig. 3 reveals that ISO STEP AP 242
“Managed Model Based 3D Engineering” contains several stand-
ardized information models that will enable 3D model-based engi-
neering identified in Fig. 2. Some elements of Fig. 3 deal with
“presentation”; these are graphical renderings that can be viewed
and understood by human supervisors and collaborators. The rest
of the contents in Fig. 3 contain representations of the information
that are machine-readable, thereby satisfying some of the key
requirements for a model-based enterprise.
ISO STEP AP 242 is developed using a modular architecture
[11]. Its modules use XML and EXPRESS schema languages, as
appropriate for the intended applications, to define the data mod-
els [12]. One of the contents of STEP AP 242 illustrated in Fig. 3
that is of interest to this paper is the 3D PMI (sometimes referred
to as product and manufacturing information), which is addressed Fig. 4 A pictorial view of the roles, dependencies, and lags in
next. the three PMI standards development processes

4 PMI defines the standardized information models for graphical presen-


tation as well as representation of PMI in the STEP specifications.
PMI is a phrase used by the computer aided design and manu- Then CAD/CAM software vendors develop import and export capa-
facturing (CAD/CAM) community to refer to geometric dimen- bilities of the STEP standards-compliant data. The software ven-
sioning and tolerancing, surface texture, finish requirements, dor implementations are tested in an implementors forum [13,14]
process notes, material specifications, welding symbols, and other to ensure that PMI have been correctly implemented and can be
annotations. Some of this information is also referred to as GPS, exchanged smoothly using the STEP standards. The process
especially in the ISO parlance. As mentioned earlier, PMI is also sequence shown in Fig. 4 is repeated when ISO TC 213 develops
expanded as product and manufacturing information, but the and issues new and updated standards.
intent still remains the same.

4.2 PMI Presentation Versus Representation. 3D graphical


4.1 PMI Standardization Processes. As noted in the intro-
presentation—including zooming, panning, and rotating a model
duction, three major ISO committees/subcommittees share the
such as the one shown in Fig. 1—of PMI is an important capabil-
bulk of the responsibility for the development of what the CAD/
ity that will enhance the human readability of complex data. Sev-
CAM community calls PMI standards. There are dependencies,
eral efforts are underway to provide light-weight viewing of PMI
and hence time lags, in their standards development processes, as
(1) using the likes of 3D PDF and JT formats, and (2) on a variety
shown in Fig. 4 and explained below.
of mobile devices. This capability will be critical in integrating
ISO TC 213 and ASME Y14.5, which is a standardization body
humans as supervisors and collaborators in a smart manufacturing
of the American Society of Mechanical Engineers, play the pri-
system. And, it will bring the benefits of smart manufacturing to
mary role of defining the syntax (symbols) and semantics (mean-
the important segment of small- and medium-sized companies.
ing of the symbols) of the PMI specifications using prose,
But 3D graphical presentation alone will not enable a manufac-
graphics, and mathematical concepts [1–3]. They also define how
turing system to become smart. We also need PMI to be machine-
these symbols should be presented on 2D drawings and, to a lim-
readable. 3D PMI representation fulfills this need and it is one of
ited extent, on 3D models. ISO TC 10 then takes over the task of
the critical capabilities that ISO STEP AP 242 brings to industry
defining the precise graphical presentation of the symbols in 2D
as a major added value over the previous AP 203 and AP 214.
and in 3D [4] for human interpretation. It is also during this period
At the minimum, 3D PMI representation can transfer PMI data
that CAD/CAM software vendors implement PMI standards in their
to various applications (for example, by automatically filling out
software and release them to the customers.
an “electronic form”) without the need for re-entry by a human
Taking input from these two ISO committees and from industry
being. This will avoid a laborious and error prone process that is
customers and CAD/CAM software vendors, ISO TC 184/SC 4 then
prevalent today. But, we can do a lot more. A CAM process plan-
ner, for example, can query the PMI representation to extract
information about machining features, datums, tolerances, and
surface texture automatically. This can then be used to choose the
machining fixtures, cutting tool, coolant, depth of cut, and feed
rate automatically. A coordinate measuring system can also query
for similar information to execute an inspection and quality con-
trol plan automatically. These measurements can be made (1)
in-process during manufacturing for real time feedback control or
(2) offline after the manufacturing to check for conformance to
specifications. Such automation will be impossible without 3D
PMI representation contained in the new ISO STEP AP 242. But
we should also be aware of its limitations that are described next.

4.3 PMI Syntax Versus Semantics. The roles of different


ISO committees and subcommittees involved in standardizing
PMI were described in Sec. 4.1. ISO TC 213 defines both the syn-
tax and the semantics of the PMI language using prose, graphics,
and mathematical principles, but these are not in machine read-
Fig. 3 An illustration of the contents of ISO STEP AP 242 (from able form and are not directly computable. This is partly remedied
Ref. [10]) by ISO TC 184/SC 4 in the form of PMI representation that

Journal of Computing and Information Science in Engineering JUNE 2015, Vol. 15 / 021001-3

Downloaded From: http://computingengineering.asmedigitalcollection.asme.org/ on 01/13/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use


captures the PMI syntax, but it does not address the PMI seman-
tics. It is important to understand this distinction in some detail.
The PMI representation in ISO STEP AP 242 (defined by ISO
TC 184/SC 4) captures the PMI syntactic structure (defined by
ISO TC 213) using EXPRESS data modeling language [15],
which is based on an entity–attribute–relationship data model.
This means that an application program that reads the PMI repre-
sentation standardized by STEP AP 242 will have all the struc-
tured data needed for algorithms, which will then interpret and
use the PMI semantics as defined by ISO TC 213 to perform man-
ufacturing and measurement tasks. So, currently, it falls upon
such application algorithms to know and interpret the semantics
of PMI. We will illustrate this with an example.
Figure 5 illustrates, as an example of PMI, how ISO TC 213
defines flatness tolerance in its ISO 1101 standard [16]. It contains
both the syntax (i.e., “indication”) and the semantics (i.e.,
“explanation”) of the flatness tolerance. The corresponding
ASME Y14.5 standard [17] defines the flatness tolerance even
more starkly as in Fig. 6. Here the syntax of the PMI language is
defined in the “this on the drawing” (which can be a on a 3D
model as well) box; its semantics is defined in the “means this”
box.
In the flatness example illustrated in Figs. 5 and 6, the PMI rep-
resentation in ISO STEP AP 242 will capture only the syntactic
structure of flatness as an attribute of a surface feature. It will be
machine readable and can be used for further computation. How-
ever, STEP AP 242 will not contain any representation of the tol- Fig. 6 Definition of the syntax and semantics of flatness toler-
erance zone, imperfectly manufactured feature, or the notion that ance by ASME Y14.5 [17]
the tolerance zone can be moved at will with respect to the manu-
factured part as long as the zone can contain the manufactured 5 Future Plans and Challenges
feature. Capturing such semantic details of PMI in a standardized
machine readable form still remains a research problem [18–21]. ISO is on track to publish the new standard ISO 10303-242 in
In the meantime, algorithms that are developed to execute manu- 2014. Portions of this standard that deal with PMI are currently
facturing and measurement tasks have assumed the responsibility being implemented and tested by major CAD software vendors.
of interpreting the semantics of PMI [22–25]. Plans are already underway to develop Edition 2 of AP 242 [10].
Meanwhile, ISO TC 213 is actively planning and publishing new
PMI standards [2,16,26,27]. With the appropriate time lag, as
indicated in Fig. 4, these PMI standards need to be included for
standardization of their machine-readable representations in future
editions of AP 242. If we label the PMI included in the current
ISO STEP AP 242 as PMI 1.0, then we should get ready for PMI
2.0 that can be an order of magnitude more complex and more
powerful than its predecessor PMI 1.0. As noted in Sec. 4.3, cap-
turing the semantics of PMI in a standardized machine readable
and computable form still remains a research problem.
The major effect of ISO STEP AP 242 PMI will be felt in
increased productivity by down-stream applications such as com-
puter aided manufacturing process planning and inspection plan-
ning. Smooth interoperability and integration of these processes
will test the belief that ISO STEP PMI representations can indeed
enable smart manufacturing systems as envisioned in Fig. 2.

6 Summary
We portrayed an upcoming ISO STEP PMI representation
standard as an enabler of what makes a manufacturing system smart.
Along the way, we described the roles and dependencies of three im-
portant ISO standards development committees and subcommittees
involved in setting PMI standards. The computer aided tolerancing
community can benefit immensely by paying close attention to the
works of all these three standards bodies, thereby positioning itself to
contribute to further digitization of manufacturing.

Acknowledgment
This paper is based on a keynote address delivered at the 13th
CIRP Computer Aided Tolerancing Conference held at Hangzhou,
China during May 11–14, 2014. We gratefully acknowledge the
Fig. 5 Definition of the syntax and semantics of flatness toler- reviews and feedbacks received at the conference that resulted in
ance by ISO TC 213 [16] the current expanded version of this paper. We also want to thank

021001-4 / Vol. 15, JUNE 2015 Transactions of the ASME

Downloaded From: http://computingengineering.asmedigitalcollection.asme.org/ on 01/13/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use


our colleagues in the ISO STEP standards community for their [13] CAx Implementor Forum, http://www.cax-if.org/
contributions to the PMI standard, and its implementation and [14] Frechette, S. P., Jones, A. T., and Fischer, B. R., 2013, “Strategy for Testing
Conformance to Geometric Dimensioning and Tolerancing Standards,” Proce-
testing. Without their hard work, the progress reported in this dia CIRP, 10, pp. 211–215.
paper would not have been possible. Any mention of commercial [15] ISO 10303-11, 2004, Industrial Automation Systems and Integration—Product
products in this paper is for information only; it does not imply Data Representation and Exchange—Part 11: Description Methods: The
recommendation or endorsement by NIST. EXPRESS Language Reference Manual, ISO, Geneva, Switzerland.
[16] ISO 1101, 2012, Geometrical Product Specifications (GPS)—Geometrical
Tolerancing—Tolerances of Form, Orientation, Location and Run-Out, Interna-
tional Organization for Standardization, Geneva, Switzerland.
References [17] ASME Y14.5, 2009, Dimensioning and Tolerancing, The American Society of
[1] Srinivasan, V., 2013, “Reflections on the Role of Science in the Evolution of Mechanical Engineers, New York.
Dimensioning and Tolerancing Standards,” Proc. Inst. Mech. Eng., Part B, [18] Barbau, R., Krima, S., Rachuri, S., Narayanan, A., Fiorentini, X., Foufou, S.,
227(1), pp. 3–11. and Sriram, R. D., 2012, “OntoSTEP: Enriching Product Model Data Using
[2] Nielsen, H. S., 2013, “Recent Developments in International Organization for Ontologies,” Comput. Aided Des., 44(6), pp. 575–590.
Standardization Geometrical Product Specification Standards and Strategic [19] Sarigecili, M. I., Roy, U., and Rachuri, S., 2014, “Interpreting the Semantics of
Plans for Future Work,” Proc. Inst. Mech. Eng., Part B, 227(5), pp. 643–649. GD&T Specifications of a Product for Tolerance Analysis,” Comput. Aided
[3] Srinivasan, V., 2008, “Standardizing the Specification, Verification, and Des., 47(2), pp. 72–84.
Exchange of Product Geometry: Research, Status, and Trends,” Comput. Aided [20] Qin, Y., Lu, W., Liu, X., Qi, Q., Zhou, L., and Jiang, X., 2014, “Ontology-
Des., 40(7), pp. 738–749. Based Semantic Interpretations of Form Tolerance Specifications in the Next-
[4] ISO 16792, 2006, Technical Product Documentation—Digital product definition Generation GPS,” 13th CIRP Conference on Computer Aided Tolerancing,
data practices, International Organization for Standardization, Geneva, Switzerland. Hangzhou, China, May 11–14.
[5] The Economist, 2012, April 21 Issue on the Third Industrial Revolution. [21] Ballu, A., Mathieu, L., and Dantan, J.-Y., 2014, “Formal Language for Geo-
[6] Advanced Manufacturing Portal, “Digital Manufacturing and Design Innovation Spelling,” 13th CIRP Conference on Computer Aided Tolerancing, Hangzhou,
(DMDI) Institute,” http://www.manufacturing.gov/docs/DMDI_overview.pdf China, May 11–14.
[7] Lubell, J., Frechette, S. P., Lipman, R. R., Proctor, F. M., Horst, J. A., Carlisle, [22] Srinivasan, V., 2007, “Computational Metrology for the Design and Manufac-
M., and Huang, P. J., 2013, “Model-Based Enterprise Summit Report,” US ture of Product Geometry: A Classification and Synthesis,” ASME J. Comput.
Department of Commerce, NIST Technical Note No. 1820. Inf. Sci. Eng., 7(1), pp. 3–9.
[8] ISO 10303-203, 2011, Industrial Automation Systems and Integration—Product [23] Srinivasan, V., Shakarji, C. M., and Morse, E. P., 2011, “On the Enduring
Data Representation and Exchange—Part 203: Application Protocol: Configu- Appeal of Least-Squares Fitting in Computational Coordinate Metrology,”
ration Controlled 3D Design of Mechanical Parts and Assemblies, International ASME J. Comput. Inf. Sci. Eng., 12(1), p. 011008.
Organization for Standardization, Geneva, Switzerland. [24] Shakarji, C. M., and Srinivasan, V., 2013, “Theory and Algorithms for
[9] ISO 10303-214, 2010, Industrial Automation Systems and Integration—Product Weighted Total Least-Squares Fitting of Lines, Planes, and Parallel Planes to
Data Representation and Exchange—Part 214: Application Protocol: Core Support Tolerancing Standards,” ASME J. Comput. Inf. Sci. Eng., 13(3),
Data for Automotive Mechanical Design Process, International Organization p. 031008.
for Standardization, Geneva, Switzerland. [25] Morse, E. P., Peng, Y., Srinivasan, V., and Shakarji, C. M., 2014, “Metrological
[10] STEP AP242 Project, http://www.ap242.org/ Challenges Introduced by New Tolerancing Standards,” ASME J. Meas. Sci.
[11] Feeney, A. B., 2002, “The STEP Modular Architecture,” ASME J. Comput. Inf. Technol., 25(6), p. 064001.
Sci. Eng., 2(2), pp. 132–135. [26] Morse, E. P., and Srinivasan, V., 2013, “Size Tolerancing Revisited: A Basic
[12] Peak, R. S., Lubell, J., Srinivasan, V., and Waterbury, S. C., 2005, “STEP, Notion and its Evolution in Standards,” Proc. Inst. Mech. Eng., Part B, 227(5),
XML, and UML: Complementary Technologies,” ASME J. Comput. Inf. Sci. pp. 662–671.
Eng., 4(4), pp. 379–390. [27] ISO/TC 213 Home Page, GPS Roadmap, http://isotc213.ds.dk

Journal of Computing and Information Science in Engineering JUNE 2015, Vol. 15 / 021001-5

Downloaded From: http://computingengineering.asmedigitalcollection.asme.org/ on 01/13/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

Vous aimerez peut-être aussi