InTech May June 2019
InTech May June 2019
A P U B L I C AT I O N O F T H E I N T E R N AT I O N A L S O C I E T Y O F A U T O M AT I O N
www.isa.org/intech
0RVW3/&VZLOOGR
EXWRQO\VRPHZLOO'RPRUH ®
Modbus®
EtherNet/IP
ASCII
&86720(55(9,(:6
“When performance and capability are mapped
against price, there is no better PLC on the
6WDUWLQJDW%;'0$5' market than the BRX. It can do everything. The
BRX is lightening fast, is easy to program, has an
ZLWK)5((6RIWZDUH extensive library of instructions, and can hold an
enormous program. More and more of my clients
The Do-more! BRX PLC series is loaded with hardware are turning to the BRX as their PLC of choice...”
and software features that provide an extensive 6FRWWLQ+XQWVYLOOH$/
toolbox to satisfy even the most demanding projects.
“The DoMore series of PLCs is the world’s best. I
• Four form factors to choose from would do nothing but DoMore-based projects if I had
• Serial/Ethernet comm ports included on most models the choice. Watch out A-B! DoMore is going to show
(with support for Modbus RTU/TCP, ASCII, EtherNet/IP, and MQTT) you how PLCs are done!k
• Hot-swappable communications port (optional) *OHQQLQ6(0,12/()/
• Built-in discrete, analog and high-speed I/O
• Advanced motion control including coordinated motion
• Internal RAM and microSD data logging
• Free PLC programming software with over 200 easy-to-use instructions,
integrated video help and a convenient project simulator www.BRXPLC.com
Kyle Shipps
Calibration Manager
PROCESS AUTOMATION
16 Industrial Process
Operation 4.0
By Martin Hollender, PhD
FACTORY AUTOMATION
The Open Process those who do not respond will indeed be left behind.
www.isa.org/InTech
DEPARTMENTS
InTech Plus is ISA’s online eNewsletter that connects
8 Automation Update automation professionals to all things automation. InTech
Industry 4.0, digitalization domi- Plus has technical content, educational training and
nate 2019 Hannover Messe, Risks videos, industry-related Q&A excerpts, and the latest and
of connected OT networks, By the greatest on industry technology and news. InTech Plus
Numbers focuses on a variety of topics, such as fundamentals of
automation and control, certification, safety, cybersecurity, the Internet of
40 Association News Things, wireless devices, human-machine interface, pressure, level, temperature,
Bassett new InTech chief editor, ISA and batch. All editorial content comes from a variety of sources, including ISA
supports Cyber Shield, ISA direc- books, training course videos, and blogs and bits from ISA’s cast of subject-
tor of strategic initiatives awarded matter experts. InTech Plus is powered by Automation.com, ISA’s premier
SANS ICS Lifetime Achievement electronic publisher of automation content. Automation professionals can
Award; certification review subscribe to InTech Plus at www.automation.com/subscribe.
Subscriptions: For ISA members, 8.65% of annual membership dues is the nondeductible portion allocated
COLUMNS to the InTech subscription. Other subscribers: $175 in North America; $235 outside North America. Multi-year
rates available on request. Single copy and back issues: $20 + shipping.
7 Talk to Me Opinions expressed or implied are those of persons or organizations contributing the information and are not to be
If disruption is the “problem,” construed as those of ISA Services Inc. or ISA.
innovation is the solution Postmaster: Send Form 3579 to InTech, 67 T.W. Alexander Drive, P.O. Box 12277, Research Triangle Park, NC
27709. Periodicals postage paid at Durham and at additional mailing office.
Printed in the U.S.A.
44 Executive Corner
Building a cybersecure manufactur- Publications mail agreement: No. 40012611. Return undeliverable Canadian addresses to P.O. Box
503, RPO West Beaver Creek, Richmond Hill, Ontario, L48 4RG
ing strategy
For permission to make copies of articles beyond that permitted by Sections 107 and 108 of U.S.
Copyright Law, contact Copyright Clearance Center at www.copyright.com. For permission to copy articles
50 The Final Say in quantity or for use in other publications, contact ISA. Articles published before 1980 may be copied for a
Beware of the hype per-copy fee of $2.50.
CHIEF EDITOR
Renee Bassett
[email protected]
If disruption is the CONTRIBUTING EDITOR
Bill Lydon
“problem,” innovation [email protected]
F
uturist Nicholas J. Webb (www.nick- creation is first an act of destruction.”
PRODUCTION EDITOR
webb.com) spoke at the 2019 Man- Creating something new usually requires Lynne Franke
ufacturing in America conference in letting go of something old—which can [email protected]
Detroit in March and discussed a number be very difficult. Transitions are pain- ART DIRECTOR
of thought-provoking trends. On the topic ful, because they destroy the status quo, Colleen Casper
of disruption, he made a statement that to pushing us beyond our comfort zones. [email protected]
me seemed fundamental, “If disruption is It is important to remember that we of-
SENIOR GRAPHIC DESIGNER
the problem, innovation is the solution.” ten do not see disruptive innovation com- Pam King
He suggested when change is occur- ing. Disruption takes us by surprise if we [email protected]
ring, people may have a tendency to are not the ones disrupting. This makes it
GRAPHIC DESIGNER
deny, hide, or be victimized by disruption. important to explore new ideas and tech-
Lisa Starck
Alternatively, people can realize change is nologies that can lead to innovative disrup- [email protected]
happening, embrace it, and benefit from tion within your manufacturing operations.
new opportunities. The task for automation professionals
is to analyze the confusing multidimen- ISA PRESIDENT
sional chaos of new technologies, expec- Paul Gruhn, PE, CFSE
tations, requirements, and processes to PUBLICATIONS VICE PRESIDENT
develop new superior solutions. Innova- Victor S. Finkel, CAP
tions may be internally complex but sim-
EDITORIAL ADVISORY BOARD
plify life for users, reducing complexity
CHAIRMAN
and increasing efficiency.
Steve Valdez
GE Sensing
Challenge
Automation professionals shine at using Joseph S. Alford PhD, PE, CAP
their experience, know-how, and creativ- Eli Lilly (retired)
ity to solve problems to improve manu- Joao Miguel Bassa
facturing and production efficiencies Independent Consultant
and quality by designing applications. Eoin Ó Riain
Another key point in his presentation Working within the limitations of exist- Read-out, Ireland
noted that “legacy” is very comfortable ing legacy systems in a plant inherently
Guilherme Rocha Lovisi
and creates a resistance to change. My limits the ability to create applications Bayer Technology Services
observation is that holding on to legacy to improve operations, productivity,
David W. Spitzer, PE
practices and systems too long can hold and profits. Newer automation systems
Spitzer and Boyes, LLC
back an organization from progress, and with superior technology give automa-
in manufacturing creates an environment tion professionals the tools to achieve Dean Ford, CAP
Westin Engineering
for competitors to overtake your business. greater results. In the environment of
Once it becomes obvious that competitors changing technology, it is important to David Hobart
Hobart Automation Engineering
are “beating” your company with sales look beyond and understand the options
and profits declining, the time, effort, for improving productivity and competi- Smitha Gogineni
and resources to become competitive are tiveness that may take new investments. Midstream & Terminal Services
expensive. At this point you are chasing Setting goals beyond today’s status quo James F. Tatera
rather than leading in your industry. and then finding ways to achieve those Tatera & Associates
goals will yield new results.
Destruction It’s hard to achieve the goal of disrup-
Thinking about innovation, the famous tive innovation if you aren’t certain what
painter Pablo Picasso said, “Every act of you are trying to accomplish. ■
S
urrounded by a bustling fair by an alliance of industry partners for equately considered in 5G standardization
illuminating the theme of “Integrated the digitization of the factory, plant, and and regulation. It encompasses stakeholder
Industry – Industrial Intelligence,” warehouse. Founders and members groups including operational technology,
the excitement and energy overflowed at endorse an open ecosystem and commit information and communication technol-
the 2019 Hannover Messe. With 6,500 to using an Open Industry 4.0 Alliance ogy, and academic groups or associations.
exhibitors from 75 countries and 215,000 Framework to achieve interoperability
attendees from 95 nations, the event that provides attractive opportunities for New wave suppliers: Since the Industry 4.0
featured Sweden as the partner country companies of all sizes. A major end goal is topic was first explored at Hannover Messe
and hosted over 80 forums and conferences to have machines in a given smart factory 2011, there has been an influx of new wave
highlighting a host of new digitalization speak the same language. exhibitors and presenters reflecting the
trends and technologies. This year’s event transition to digitalization. Hannover Messe
revealed a number of organizations focused 5G Alliance for Connected Industries and 2019 continued to see the traditional in-
on providing standards and information Automation (5G-ACIA): This global forum, dustrial suppliers exhibit and show off new
to assist the modernization and digital prominently seen throughout Hannover products, but there has been a change. The
transformation of manufacturers. Messe 2019 halls, was formed to address, booths of high-technology companies—
discuss, and evaluate the technical, regula- including Amazon, Microsoft, Oracle, SAP,
Open Industry 4.0 Alliance: Some of the tory, and business aspects of fifth-genera- IBM, Dell, Hewlett Packard Enterprise, Hua-
biggest news was the announcement of The tion (5G) cellular network technology for wei, Dassault Systèmes, CISCO, and PTC—
Open Industry 4.0 Alliance, a collaboration industrial applications. The dominant objec- have gotten bigger and more prominent.
with the goal of overcoming proprietary tive of 5G-ACIA is the best possible applica- New high-tech companies, either upstarts
solutions. The charter of the organization bility of this wireless communications tech- or those crossing over from computer/IoT
is to create customer value through holistic nology for connected industries. 5G-ACIA applications, are also increasingly bringing
interoperable Industry 4.0 solutions and works to ensure that the interests and spe- value to manufacturing. ■
services in a common framework powered cific aspects of the industrial domain are ad- – By Bill Lydon, InTech
T
ime to market is an essential competitive edge in a digi- strategy. The following examples show vulnerabilities in today’s
tal marketplace that has placed growing pressure on the transportation systems.
rapid delivery of goods. In addition to developing new Satellite/Internet communications: Navigation and cargo sys-
products and services, success today depends upon new func- tems are increasingly connected to satellite and Internet commu-
tions like DevSecOps teams and agile software development, nication, escalating cyberrisk. In 2013, a University of Austin stu-
more speed and bandwidth, on-demand infrastructures span- dent participating in a sanctioned experiment aboard an $80M
ning multicloud environments to manage big data, and hyper- yacht spoofed the navigation system, steering the ship off course.
connectivity across data and resources. Advanced persistent threats: APTs are a clear and present dan-
Few places are experiencing more of the cyberimpact of this ger to the transportation industry. Compromised ticketing and
new business model than the operational technology (OT) trans- scheduling systems, for example, can shut down transportation
portation sector. Organized cybercriminals have actively exploit- hubs. Airlines security experts agree that more intelligence across
ed container shipping companies and container port operators. the cyber kill-chain must be shared between carriers, but this
By hacking Internet-connected OT systems, cybercriminals have requires public-private cooperation that does not currently exist.
accessed ICS-based cargo systems to redirect containers or make Phishing attacks: The impact can be severe when cybercrimi-
them disappear off the grid entirely. They access aircraft systems nals execute a masquerade attack and gain network privilege,
by breaking into Internet-connected OT subsystems such as but the attacks are difficult to execute. Once a cyberattacker has
communication, maintenance, catering, and baggage handling. access, the threat can extend to IT resources.
Although OT systems do not present the sort of personal data Espionage: According to one report, 47 percent of malware
value that many traditional cybercriminals seek, targeting criti- aimed at manufacturers was intended to steal intellectual property.
cal infrastructure still has huge appeal. Cybercriminal agendas According to the National Center for Manufacturing Sciences, 21
include holding a critical system hostage, manipulating a stock percent of manufacturers lost intellectual property as a result. ■
price, or even operating as a cyber “hit man” as a competitive – By Rick Peters, Fortinet
At its annual Innova- another record. Endress+Hauser owns digital communications, and instrument
By Dave Emerson
FAST FORWARD
● OPAF, under the guidance of The Open
Group, has developed the O-PAS Standard, Version 1.0.
● The standard defines minimum requirements for components that can
be used to create federated process automation systems with an open
and interoperable reference architecture.
● O-PAS Version 1.0 was released in January 2019 as a preliminary
standard of The Open Group; OPAF will incorporate industry
feedback after an interoperability workshop this June.
P
rocess automation end users and sup- Figure 1. A broad sampling of suppliers and end
pliers have expressed interest in a stan- users are highly interested in the scope of the
OPAS under development by OPAF, because it
dard that will make the industry much
touches on all the key components of industrial
more open and modular. In response, the Open automation systems: hardware (I/O), the com-
Process Automation™ Forum (OPAF) has munication network, system software (e.g., run
worked diligently at this task since November time, namespace), application software, and the
Business
2016 to develop process automation standards. data model.
systems
The scope of the initiative is wide-reaching, as
it aims to address the issues associated with the
process automation systems found in most Open process
Advanced control,
industrial automation plants and facilities to- manufacturing automation
day (figure 1). execution system scope
(MES)
It is easy to see why a variety of end users and
suppliers are involved in the project, because
the following systems are affected: Distributed control Programmable logic Safety
system (DCS) + control (PLC) + instrumented
● manufacturing execution system (MES)
human-machine human-machine system (SIS)
● distributed control system (DCS) interface (HMI) interface (HMI)
● human-machine interface (HMI)
Standard of standards has been identified by end users as a will develop conformance tests to make
Creating a “standard of standards” for requirement to allow their intellectual sure products interoperate properly. The
open, interoperable, and secure auto- property (IP), in the form of control O-PAS Version 1 profiles are:
mation is a complex undertaking. OPAF strategies, to be portable. Existing stan- ● Level 1 Interoperability Hardware
intends to speed up the process by le- dards under evaluation for use in Ver- Profile: A certified product claiming
veraging the valuable work of various sion 2.0 include: conformance to this profile shall
groups in a confederated manner. ● IEC 61131-3 for control functions implement OSM-Redfish.
The OPAS Standard will reference ex- ● IEC 16499 for execution coordination ● Level 2 Interoperability Hardware
isting and applicable standards where ● IEC 61804 for function blocks Profile: A certified product claiming
possible. Where gaps are identified, O-PAS Version 3.0 will address ap- conformance to this profile shall
OPAF will work with associated or- plication portability, which is the abil- implement OSM-Redfish BMC.
ganizations to update the underlying ity to take applications purchased from ● Level 1 Interoperability Software Pro-
standard or add OPAS requirements to software suppliers and move them file: Software claiming conformance
achieve proper definition. Therefore, among systems within a company in to this profile shall implement OCF-
OPAF has already established liaison accordance with applicable licenses. 001: OPC UA Client/Server Profile.
agreements with the following organi- This release will also include the first ● Level 2 Interoperability Software Pro-
zations: specifications for hardware interfaces. file: Software claiming conformance
● Control System Integrators Associa- to this profile shall implement OCF-
tion (CSIA) Under the OPAS hood 002: OPC UA Client/Server and Pub/
● Distributed Management Task Force The five parts that make up O-PAS Ver- Sub Profile.
(DMTF), specifically for the Redfish sion 1.0 are listed below with a brief The term “Level” in the profile names
API summary of how compliance will be refers to profile levels.
● FieldComm Group verified (if applicable):
● Industrial Internet Consortium (IIC) ● Part 1 – Technical Architecture Over- Part 4 – Connectivity Framework (OCF)
● International Society of Automation view (informative) forms the interoperable core of the
(ISA) ● Part 2 – Security (informative) system. The OCF is more than just a
● NAMUR ● Part 3 – Profiles network, it is the underlying structure
● OPC Foundation ● Part 4 – Connectivity Framework (OCF) allowing disparate components to
● PLCopen ● Part 5 – System Management interoperate as a system. The OCF will
● ZVEI use OPC UA for OPAS Versions 1.0, 2.0,
Part 1 – Technical Architecture Overview and 3.0.
Additionally, OPAF is in discussions (informative) describes an OPAS-confor-
with AutomationML and the ISA Secu- mant system through a set of interfaces Part 5 – System Management covers
rity Compliance Institute (ISCI) as an to the components. Read this section foundational functionality and inter-
ISA 62443 validation authority. In addi- to understand the technical approach face standards to allow the manage-
tion to these groups, the OPC Founda- OPAF is following to create the O-PAS. ment and monitoring of components
tion has joined OPAF as a member, so using a common interface. This part
no liaison agreement is required. Part 2 – Security (informative) addresses will address hardware, operating sys-
As an example of this cooperation in the necessary cybersecurity functional- tems and platform software, applica-
practice, OPAS Version 1.0 was created ity of components that are conformant tions, and networks—although at this
with significant input from three exist- to OPAS. It is important to point out point Version 1.0 only addresses hard-
ing standards, including: that security is built into the standard ware management.
● ANSI/ISA 62443 (adopted by IEC as and permeates it, as opposed to being Conformance criteria are identified
IEC 62443) for security bolted on as an afterthought. This part by the verb “shall” within the O-PAS
● OPC UA adopted by IEC as IEC 62541 of the standard is an explanation of the text. An OPAF committee is working on
for connectivity security principles and guidelines that a conformance guide document that
● DMTF Redfish for systems manage- are built into the interfaces. More spe- will be published later this year, which
ment (see www.dmtf.org/standards/ cific security requirements are detailed explains the conformance program
redfish) in normative parts of the standards. The and requirements for suppliers to ob-
detailed normative interface specifica- tain a certification of conformance.
Next step: Configuration portability tions are defined in Parts 3, 4, and 5.
Configuration portability, now under These parts also contain the associated Technical architecture
development for OPAS Version 2.0, will conformance criteria. The OPAS Standard supports commu-
address the requirement to move con- nication interactions that are required
trol strategies among different auto- Part 3 – Profiles defines sets of hardware within a service-oriented architecture
mation components and systems. This and software interfaces for which OPAF (SOA) for automation systems by out-
External OT
On-premise OT data center Enterprise data center
(executing IEC 62264 Level 2 and 3 functions) IT data centers (executing IEC
(executing IEC 62264 62264 Level 2
Level 4 functions) External data centers may run
and 3 functions) DCFs that are connected to the
Advanced computing platform
OCF through a firewall.
DCF DCF DCF Business platform Non-O-PAS DCF Stand-alone DCF environments
Transactional environments
computing platform DCF may be used for functions such
as offline engineering and
simulation.
Business platform
communicates through Legend
apps running in a DCF,
DCN Non-O-PAS
not directly to the OCF
DCN conformant
DCN = DCP + DCFs platform
Connectivity framework (OCF) O-PAS
component
DCN
DCN
DCN
DCN
DCN
DCN
DCN
DCN
lining the specific interfaces the hard- and also allows it to run control software. Advanced computing platform (ACP):
ware and software components will An ACP is a computing platform that
use. These components will be used to Distributed control platform (DCP): A implements DCN functionality but has
architect, build, and start up automa- DCP is the hardware and standard soft- scalable computing resources (mem-
tion systems for end users. ware interfaces required in all DCNs. ory, disk, CPU cores) to handle appli-
The vision for the OPAS Standard is to The standard software interfaces are cations or services that require more
allow the interfaces to be used in an un- a common platform on top of which resources than are typically available
limited number of architectures, there- control software programs run. This on a small profile DCP. ACPs may also
by enabling each process automation provides the physical infrastructure and be used for applications that cannot be
system to be “fit for purpose” to meet interchangeability capability so end easily or efficiently distributed. ACPs
specific objectives. The standard will not users can control software and hard- are envisioned to be installed within
define a system architecture, but it will ware from multiple suppliers. on-premise servers or clouds.
use examples to illustrate how the com- Within the OPAS Standard, DCNs
ponent-level interfaces are intended to Distributed control framework (DCF): represent a fundamental computing
be used. System architectures (figure 2) A DCF is the standard set of software in- building block (figure 3). They may be
contain the following elements: terfaces that provides an environment hardware or virtual (when virtual they
for executing applications, such as con- are shown as a DCF as in figure 2), big or
Distributed control node (DCN): A trol software. The DCF is a layer on top of small, with no I/O or various amounts. At
DCN is expected to be a microproces- the DCP that provides applications with the moment, allowable I/O density per
sor-based controller, I/O, or gateway a consistent set of O-PAS related func- DCN is not settled, so some standardiza-
device that can handle inputs and out- tions no matter which DCN they run in. tion in conjunction with the market may
puts and computing functions. A key This is important for creating an efficient drive the final configuration.
feature of O-PAS is that hardware and marketplace for O-PAS applications. DCNs also act as a gateway to other
control software are decoupled. So, the networks or systems, such as legacy
exact function of any single DCN is up OPAS connectivity framework (OCF): systems, wireless gateways, digital
to the system architect. A DCN consists The OCF is a royalty-free, secure, and field networks, I/O, and controllers
of hardware and some system software interoperable communication frame- like DCS or PLC systems. Industrial In-
that enables the DCN to communicate work specification. In O-PAS Version 1, ternet of Things (IIoT) devices can also
on the O-PAS network, called the OCF, the OCF uses OPC UA. be accessed via any of these systems.
kelleramerica.com
oem pressure transmitters
Industrial Process
Operation 4.0
D
By Martin igitalization, Internet of Things (IoT), for the Industry 4.0 high-performance work force.
Hollender, PhD big data, artificial intelligence, and vir- Work environments must support collaboration
tual reality are some examples of rapidly at all levels and support high-performance work
developing areas of technology that will have a around the clock.
big impact on how industrial processes will Like the situation in the transport sector with
operate in the future. Normal operations that the advent of self-driving cars, the way industrial
are already highly automated will be even more processes operate is dramatically changing.
automated in the future. Tasks like fault detec- Today’s sophisticated digital automation pro-
tion, diagnosis, and process optimization are grams can handle most situations. Cheap sensors
becoming more complex. connected to powerful artificial intelligence algo-
Many of those tasks are best handled by in- rithms, like image recognition or vibration moni-
terdisciplinary teams with broad expertise and toring, can increasingly replace human sensing. A
knowledge about process, plant, operations, single operator can take responsibility for larger
maintenance, networks, sensors, and actuators. and larger plant sections.
Collaborative process operations make it pos- Integrated industrial information systems gath-
sible to efficiently bring disciplines together to er operational data to enable collaboration across
focus on the problem at hand. Big data and arti- locations, disciplines, and organizations. They
ficial intelligence tools support teams and make make real-time data easily available to the appro-
them as efficient as possible. Previously isolated priate individuals.
control rooms become networked control centers However, the reality is often far from ideal. In a
FAST FORWARD
● Digitalization, IoT, big data, artificial intelligence, and virtual reality
will improve industrial process operation.
● Work environments must support collaboration at all levels and
support high-performance work around the clock.
● Knowledge workers in process operation still spend too much time
searching for data in information silos or proprietary tools.
case example about offshore platforms, McKinsey motely, as is the case for many offshore platforms.
has shown that although huge amounts of data Modern control rooms have turned into net-
are already being collected, only a small portion is worked information and communication centers
actually being used as a basis of operational deci- where collaboration workflows come together.
sions (figure 1). This is currently changing, as the The remaining operators need a supportive work
technical infrastructure that allows data-driven environment that helps them stay vigilant and
decision making becomes available. carry out their jobs as effectively as possible.
Other important trends include flexible modu-
lar plants for producing small quantities of fre- Breaking down information silos
quently changing products. Such processes are Modern process plants are complex and highly
more difficult to operate because of the frequent coupled systems. As a result, a problem in one part
product changeovers, and it is more difficult to of the process will tend to propagate across dif-
gather experience. ferent subsystem and plant components. The ad-
New big data and artificial intelligence methods vanced automation systems in use also add com-
can predict upcoming problems long before they plex dynamic interactions between the different
affect production. They also enable prescriptive plant components, making it difficult to obtain a
maintenance strategies. Remote operation is be- clear assessment of a potential problem. Collab-
coming more widely used. Often it makes sense orative efforts from a multidisciplinary team are
to bring in highly specialized remote expertise. needed to effectively troubleshoot, diagnose, or
Sometimes even the whole plant is operated re- optimize process dynamics. In addition, the
From a base of 30,000 data tags, close to zero tags are used to inform operational decisions. Although industrial applications
Comment Source have been lagging behind consumer
People and 0% Schedule predominately based on OEM- and enterprise solutions, industry is
processes recommended maintenance intervals now catching up to provide the same
Interviews with
operational staff
Deployment <1% No interface in place to enable real-time level of digital support to the industrial
analytics to “reach” offshore worker and the office worker, whether
Analytics <1% Reporting limited to a few KPIs that BI and KPI they are in the control room, in the
are monitored in retrospect walkthrough
plant, or in a remote location (figure 2).
Data ~1% Data cannot be accessed in real time, Information previously hidden within
management enabling only ad hoc analysis Walkthrough of infra-
structure and band- the control systems or proprietary tools
width between off-
Infrastructure 60% Only about 1% can be streamed onshore and onshore is now increasingly made available
for day-to-day use
through improved connectivity and inte-
Data capture About 40% of all data is never stored— Assessment of storage gration across different systems and net-
100% capacity (on the
the remainder is stored locally offshore highest-capacity asset) work layers. Web-based applications are
~30,000 tags measured available to support the consolidation
of data from different systems and tools,
Figure 1. Case example about data-driven decisions Source: McKinsey making these easily accessible from one
place. Easy data access and a common
work environment is the first step for ef-
fective collaboration to support process
operation. Improvements in analytics
and visualization techniques also help
workers make sense of the increasing
amount of data available.
Other technology trends are also sup-
porting a new collaborative approach to
working. After many years of teething
troubles, video conferencing technol-
ogy has matured and is moving from a
nice-to-have technology to a necessity.
Several companies now have remote
operation centers that support the local
control rooms with continuously open
video links between locations. High-
Figure 2. Although industrial applications have been lagging behind consumer and enter- quality video conferencing technology
prise solutions, industry is now catching up to provide the same level of digital support to is also available from mobile devices or
the industrial worker and the office worker. personal workstations, so operators can
get instant access to remote expertise
via video conferencing whenever they
highly advanced systems used to sup- port multidisciplinary collaboration, need it. In combination, the introduc-
port plant operations may also require and therefore tend to execute work tion of digital technology for easy ac-
the involvement of specialized exper- based on a relay race approach instead cess to information, independent of
tise, often represented by an external of as a collaborative effort. location, and the proliferation of video
supplier. However, industrial companies are conferencing to support remote col-
Unfortunately, collaboration be- realizing that they need to improve the laboration, are blurring the boundaries
tween personnel from different disci- way they work to stay competitive in an between local and remote operation.
plines, locations, and organizational increasingly volatile market. The digitali- Modern automation systems cover
boundaries is often hindered by the fact zation trend is sweeping across the indus- most aspects of normal operation
that the information needed to solve tries. Companies are taking actions to but also handle many abnormal situ-
the problem at hand is hidden within improve workforce effectiveness through ations. Advanced control techniques,
numerous information silos. Knowl- the introduction of digital technolo- such as model-predictive control
edge workers in process operation still gies. Many companies are introducing (MPC) and state-based control, allow
spend too much time searching for “bring-your-own-device” policies and the automation of very complex tasks,
data in information silos or proprietary deploying solutions so their employees such as the startup of a plant. Auto-
tools. Many companies also lack the or- can work effectively wherever they are: at matic control performs better than
ganization and work processes to sup- the office, on trips, or from home. typical human operators. The opera-
tor is less and less involved in the inner was the incorrect interpretation of avail- to achieve best results. Many of the tasks
control loops with direct contact to the able measurements. Quite likely, with either can be performed by centralized
process. The tasks shift more and more advice from highly qualified remote ex- internal service centers or can be out-
to supervisory control, where the op- perts, the company would have avoided sourced to specialized external service
erator manages and supervises a large this accident. providers.
number of control modules. The high complexity of modern plants Typical goals are increased through-
requires expertise from many different put, efficiency, and uptime for the pro-
Bridging the knowledge gap domains (e.g., MPC, chemistry, electri- duction plant. These goals are accom-
But being less involved in direct process cal drives, distributed control systems). plished by a structured approach to
control also means fewer opportunities It is impossible for most plants to hire revealing the sources of process varia-
to develop a feeling for the process by personnel with sufficient knowledge in tions and upsets and how they are cur-
training on the job. (This problem was
dramatically illustrated with the ac-
cident of flight AF447. The autopilot To be able to take over when automation fails, operators
discovered inconsistent speed mea-
surements from all three redundant
need higher qualifications and a profound understanding
speed measurements and switched of the technical process, the automation system, and the
into manual mode. The pilot did not
have enough experience flying at great
control modules.
heights and was overburdened with
this sudden and unexpected transfer all these areas. Modern collaborative rently handled. By reducing process
of responsibility. He went into climb environments make it possible to bring variations, organizations will increase
mode, which reduced the speed of the in remote expertise as needed. the operational flexibility, plant regular-
plane and finally led to the crash.) Managing key performance indica- ity, safety, and integrity, while reducing
To be able to take over when auto- tors for process operations in areas such off-spec production, energy costs, envi-
mation fails, operators need higher as control loop performance, alarm ronmental impacts, operator stress, and
qualifications and a profound under- management, energy efficiency, and equipment wear.
standing of the technical process, the overall equipment efficiency is not a For example, Dow Chemical intro-
automation system, and the control classic operator task but is becoming duced a global analytics layer that turns
modules. Simulator training is neces- more and more important to ensure vast amounts of data into information
sary to develop a feeling for the pro- good production performance. Disci- and metrics anyone could see. Experts
cess. Modern operators should also be plines such as operations, maintenance, from a centralized Analytical Technology
deeply involved in the optimization of and analytics need to go hand in hand Center can now support plants globally
process operations, because such an
activity keeps them involved and helps Phase 3
to build up the required knowledge Sustain
that allows them to take over in case of 100% (ProcessPRO)
automation failure. Phase 1
Another area where Industry 4.0 will Diagnose Ideal
(fingerprints) Phase 2 Performance
Process performance potential
It
By Duncan Micklem seems that every software initiative in The process industries saw this and feared they
the energy and chemical industry is would be left behind.
now tagged with the name “digitaliza- In the consumer world, entrepreneurship
tion.” Yet similar initiatives a few years ago were caused the digital revolution. Consumers ben-
called Six Sigma or lean manufacturing proj- efited from improved efficiency and conve-
ects. In essence, they are all focused on opera- nience, greater social connectivity and personal
tional excellence, but digitalization has risen to security, and even elevated status. Minor im-
the top of the C-suite agenda due to the rapid provements in a consumer’s experiences have
penetration of new technologies disrupting the changed suppliers’ business models massively
way consumers are buying and using everyday and forever—newcomers have entered from
products and services. nowhere, and some household names that
Music and movies are provided as services failed to respond are gone forever (think Block-
that learn about your preferences and those of buster, Sears, and Toys“R”Us). Some of our kids
your peers; they used to be sold as a hardware are growing up knowing nothing different.
product that you bought and owned. More and The response of the energy and chemical in-
more, people are buying products of all kinds dustry is digitalization. In this world, the gains
from unknown vendors in far-off locations to be had are much more valuable—greater
through trusted online marketplaces. They are profitability, improved asset performance, and
delivered to your doorstep, so you no longer better competitiveness. We can expect much
have to find and visit a local vendor and hope it more drastic consequences—hundreds of mil-
has stock on hand. Newspapers and magazines lions of dollars are at stake, and those who
are failing as we consume news and entertain- embrace digitalization will prosper. Those who
ment in real-time through our phones, and do not respond will indeed be left behind: either
often for free. Our cars are mobile information consumed by those who do or fatally destroyed
centers with the intelligence to save us from our in the marketplace.
own bad driving habits. And we can control our So digitalization is more than just another Six
homes and maintain their security remotely. Sigma or lean manufacturing project. It is an
Digitalization
journey
All digitalization initiatives
go through this series
of five steps.
Frost & Sullivan estimated that process Analytics are necessary for foresight and oversight, and
industries use less than 5 percent of the we also consider them beneficial for hindsight and insight.
data that is collected—95 percent of the We are strong believers in using first principles-based ana-
data is either siloed (used selectively), lytics tools in conjunction with emerging correlation-based
dark (unused), or not consistently in analytics (also known as statistical or stochastic analytics)
READINESS
use. Problems of assigning context to for situational awareness—a so-called “ensemble approach.”
data and poor quality have also been identified. First principles tools bring rigor due to their built-in under-
To be ready for digitalization, the impediments to data standing of physics, chemistry, and dynamics, but at the cost
utilization must be addressed: (a) data readiness (data suf- of complexity and relatively high computation time. Corre-
ficiency, data trust, data propagation, and data governance), lation-based analytics suffer from lower fidelity without any
(b) infrastructure readiness (physical infrastructure; secu- guarantee of feasibility, but with the advantage of simplicity
rity, privacy, and confidentiality; software infrastructure and and speed of solution.
cloud infrastructure), (c) consumption readiness, and (d)
In the same way that we recommend an
people readiness. A wise approach is to perform a readiness
ensemble approach to situational aware-
assessment and to tackle any readiness issues before starting
ness, we also believe that decision mak-
(or perhaps in parallel with) a digitalization initiative.
ing should be grounded on first princi-
ples in conjunction with correlation-based
To take the right actions to improve a DECISION tools as necessary. Decision making is
plant’s operation, it is important to un- MAKING about looking for answers. In an operat-
derstand the potential for improve- ing plant seeking to improve performance, there are three
ment. Situational awareness is therefore main kinds of answer that can be sought: (a) forecasting (“what
a crucial step—knowing how the plant next?”)—a judgment of what is likely to happen in the future
SITUATIONAL is and has been performing in absolute based on knowledge of the past; (b) prediction (“what if?”)—
AWARENESS
terms (“hindsight”), understanding an estimate of what will happen in the future based on chang-
where it has capacity for improvement versus its constraints es that could be made in the present; and (c) optimization
and optimal capability (“insight”), predicting responses to (“what is best?”)—an approach that answers the question, “Of
changes (“foresight”), and assessing the success and value all possible changes that can be made, which has the best eco-
of such changes (“oversight”). nomic outcome?”
Tools associated with hindsight and insight are largely In the energy and chemical industry, there are many complex
visual—dashboards, BI tools, spreadsheets. These gain sig- decisions to be made due to the vast number of variables that
nificant value when they align with goals, targets, and con- can be controlled and the large quantity of disturbances and
straints. Therefore, to present decision makers with valid in- constraints. Correlation-based decision tools are useful when
formation in dashboards, for example, the right tools must be accuracy is not as important as feasibility and when the answer
applied to each situation being analyzed. lies within an already-experienced operating window. However,
sloppiness in accuracy comes at a cost—the actual optimal so- value of the applications.
lution is likely worth a lot more than a simply feasible solution. Achieving this involves goal monitoring and economic
Rigorous models will always find the best answer. Always. stewardship, knowledge management, management of
change, and a value-versus-cost mindset around technology
Being ready, situationally aware, and mak- and capability refresh.
ing the right decisions only guarantee suc- No doubt this will have a significant cost. However, the value
cess with efficient and effective operation- that it unlocks will be orders of magnitude greater, enabling
al execution. Digitalization compresses sustainment actions to pay for themselves many times over.
time horizons, which means not only do- Our organization recently worked with a major multi-
OPERATIONAL ing the same thing faster, but becoming national oil company that was looking for a flexible, light-
EXECUTION
liberated to do completely new things. weight, and cloud-based solution to manage operating goals
The more encompassing the decision, the longer it takes to and constraints at its western Canada plants. It wanted to en-
make and the more economically and organizationally im- sure its assets were always operated according to best prac-
pactful it is, and for a longer time. A wise business decision tice, something it had struggled with due to a retiring skilled
may reap rewards for years, whereas a poor business decision workforce being replaced by younger, less experienced staff.
may have long-term costly consequences. Embracing the digitalization concept, the oil company
Automation, on the other hand, by its inherent nature, worked with our organization to adopt Operating Goals Man-
makes decisions very quickly based on very recent limited ager™ (OGM) as a standard application delivered through a
data. The scope is typically much more contained, and the software-as-a-service model. The application allowed users
automation actions can be suspended or terminated quickly. to define measured variables and tasks as indicators. Each
So, the more informed and timelier the decision process be- of these indicators contained company knowledge, such as
comes, the more likely the decision will be good, the quicker reasons for the specific operating envelope; consequences
it will be to execute, and the easier it will be to course correct. for not addressing an excursion in a timely fashion; recom-
Digitalization accelerates information flow, increases the mended actions to address an excursion; and stored com-
power of analytics, and automates much of the execution, pany procedures and documents.
which greatly condenses the decision/execution time horizons, As a result, OGM is being used in four gas plants in west-
allowing strategic business decisions to be made in real time, ern Canada as a cloud service, tracking thousands of live in-
and the results to be visible and available almost immediately. dicators in real time and supporting hundreds of users. This
The tools of operational execution (advice-based open- is just one example of many where a digitalization strategy
loop actions, closed-loop control, procedural automation, has helped improve overall operational excellence.
and closed-loop optimization) start with best practices.
Replicating poor or average business processes in a digi- Do not get left behind
tal environment does not ensure delivery of superior results. Digitalization creates and sustains competitive advantage and is
The focus should be on acquiring best practices to execute one of the key strategies a plant can adopt in pursuit of opera-
the organization’s work, and thereafter on finding the digital tional excellence. Despite this, many in the industry still remain
means to institutionalize automation of as much of each of confused or irritated by digitalization. Some feel it is merely an in-
these processes as possible. formation technology issue and do not understand its relevance
for operations; some are frustrated by the plethora of buzzwords;
Digitalization is typically not a one-time some see it as hype and fail to see the value proposition.
hit, where the benefits are achieved and Digitalization leaders, however, see it as a holistic busi-
stay forever. Unless proactive steps are tak- ness issue and are already making huge strides forward in
en, benefits will almost always decline over productivity, efficiency, flexibility, and agility. Those who
time, and the opportunity to capture in- do not realize the value digitalization has to offer risk being
VALUE cremental benefits will diminish too. This left behind. Failure to adapt and transform means that the
SUSTAINMENT happens for a number of reasons: magnitude of value being lost will continue to increase—the
● the economic basis for the solution changes digitally wise will consume the laggards in the market. ■
● goals change
ABOUT THE AUTHOR
● plant performance changes
Duncan Micklem ([email protected]) is the executive vice
● business priorities change
president, strategy and marketing at KBC, a Yokogawa com-
● focus by people (management, engineers, operators) changes
pany. Based in Houston, Texas, Micklem’s responsibilities include
● technology changes.
business planning, M&A/alliances, investment prioritization/cap-
Our approach to sustaining digitalization value entails go-
ital allocation, and marketing. Micklem has a degree in biology
ing above and beyond compliance with how new digital ap-
and geography from the University of Exeter, U.K., and an MBA
plications are implemented. It is an approach where there
from Cass Business School, U.K.
is a clear sense of ownership by the organization, especially
the front-line operators, through recognition of the added View the online version at www.isa.org/intech/20190603.
Can reconfigurable
computing be the
answer to our woes?
By Gineesh Sukumaran
A
lthough machine vision may seem like a once took several minutes to compute images
new concept, we can trace its origins to are now expected to do the same computation in
the 1960s. Back then, machine vision seconds (real-time computing). A prime example
existed as raw image files. A paradigm shift hap- of this is bottle inspections in a soda manufac-
pened with the advent of digital photography. turing facility. Soda companies use an advanced
Slowly, companies began developing a software four-camera system that can inspect a whopping
ecosystem that could import these digital photo- 72,000 bottles per hour.
graphs and process them in multiple ways.
Alongside superior cameras and greater image Why does real-time machine vision need
processing capability came automation that en- more flexibility?
abled machines to independently capture images Engineering real-time machine vision is an ex-
without human intervention. Today, advanced tremely expensive process. Several companies
artificial intelligence (AI) and deep learning tools have accomplished it by either building dedicated
have taken this one step further by augmenting field-programmable gate array (FPGA) systems or
machines with the intelligence to make right deci- installing custom camera systems that have prefed,
sions based on the images captured. application-specific integrated circuits (ASICs).
As machine hardware on the shop floor goes The limiting factor here is the need for a dedicated
through transformational upgrades, manufactur- design architecture for every single application.
ers want the ancillary things around the hardware This is expensive and not scalable.
to keep pace. Hence, machine vision systems that Going back to the soda bottle inspection system,
every camera comes fitted with a dedicated ASIC puting, the camera system can be trained to ac-
chip. These high-end cameras cannot be repur- commodate modifications to the packaging or the
posed for any other application. To make matters inspection process.
worse, inspection processes and packaging in
manufacturing facilities are modified from time Ideal reconfigurable computing platform
to time. Tweaking the vision inspection system to Whatever the approach may be, configuration
work with the new processes or packaging requires management and development systems are inte-
extensive and expensive modifications. gral parts of a reconfigurable computing platform.
Given these limitations, machine vision sys- There are a variety of choices when considering
tems must be designed so they can be reused for the development of a reconfigurable system. Some
multiple applications and easily modified to meet of the critical elements to bear in mind when im-
changing product designs, manufacturing pro- plementing a task on reconfigurable systems are:
cesses, varying components, and business needs. ● Overhead for run-time reconfiguration: How
much additional work needs to be done to use
Reconfigurable computing is the best bet the flexibility offered by the reconfigurable
Selection of computing platforms for automa- systems?
tion tasks is often governed by the trade-off ● Configuration granularity: To what level can an
between two important parameters: efficiency algorithm be decomposed during the imple-
and flexibility. The search for a balance between
the approaches brings us to reconfigurable
computing. Reconfigurable systems are imple- Processor RC ASIC
mented with programmable logic. The best part
is that hardware circuits can be easily upgraded Program Configuration
memory
to meet the demands of fresh or modified vision
applications.
Processors with a fixed architecture allow the
Data
Data
Data
Processor RC device ASIC
user to temporarily compose the operations pro-
vided by the arithmetic logic unit (ALU). ASIC
processors are implemented by spatially compos-
ing dedicated functional units. Reconfigurable Concept of reconfigurable computing
computing uses both temporal and spatial com-
position—that is, these systems typically have
a programmable hardware logic and a group of Application
programmable interconnection networks. There
are several approaches available to accomplish a
reconfigurable computing system: Development system
● fine-grained systems: FPGAs and logic gates Software environment
HDL
as programmable elements, connected by
switches and wires Compiler
Configuration management
● coarse-grained systems: software-configu-
CAD tools
rable processors and a programmable net-
work on chips Reconfigurable computing engine
● heterogeneous reconfigurable systems: a
combination of fine-grained and coarse-
Composition of a reconfigurable system
grained elements
The ramifications are awesome. Take a tradi-
tional packaging vision system that comes with
FAST FORWARD
a prefed ASIC in a dedicated camera setup. If the ● Legacy vision systems are limited by the need for a dedicated design
product packaging were changed, new patterns architecture for every application. This is expensive and not scalable.
and components were added, or the text on the ● We can now deploy reconfigurable machine vision systems that are reus-
package was modified, we would need a new ASIC able for multiple applications and can be easily modified for changing
chip for every production line in every manufac- product designs, manufacturing processes, varying components, and
turing plant of this company. The costs and time business needs.
involved in design, fabrication, testing, and roll- ● Advanced AI and deep learning tools have taken this one step further
out are massive. by augmenting machines with the intelligence to make right decisions
based on the images captured.
On the other hand, with reconfigurable com-
mentation—the number of partitions ● Low-level vision algorithms: These are candidates for reconfigurable architec-
and their size—which mainly depends simple algorithms that typically oper- tures. It takes a great deal of effort to
on the resources available on the re- ate on a small neighborhood. Low- make reconfigurable processors work
configurable processor. level vision algorithms are easy to with midlevel vision algorithms.
● Ease of use of SW and development implement and provide great compu- ● High-level vision algorithms: These
system: This largely depends on how tational efficiency when operating on algorithms are computationally in-
the manufacturer packages different a large amount of data. Some typical tensive, nonlocal, nondeterministic,
hardware and software elements and examples include filtering operations, and extremely complex to process.
provides them to the developer. image enhancement, denoising, and A large majority of the matching, ob-
Based on the specific requirements at image compression. When processing ject recognition, classification algo-
hand, the most appropriate elements can low-level algorithms, reconfigurable rithms, decision making, inferences,
be put together to design an application. systems are extremely efficient—up and retrieve queries from database
to 100 times faster and more efficient algorithms fall under this category. As
Optimization of vision algorithms than CPU-based vision systems. the number of parallel operations are
The most important aspect when work- ● Midlevel vision algorithms: Though high, one can implement extremely
ing on a reconfigurable system has to they operate on a small neighbor- efficient and flexible systems using re-
do with programming the logic to the hood, midlevel vision algorithms are configurable processors.
reconfigurable processor. This requires more complex and often require ran-
an understanding of both the platform dom memory access. Some examples Reconfigurable computing pros/cons
and industry domain. of midlevel vision algorithms include The advantages
Every machine vision system can be connected component labeling and ● Increased efficiency: One of the greatest
considered a pipeline that uses three segmentation and feature extraction advantages of reconfigurable systems
kinds of algorithms, low, mid, and high, techniques. The components that can is the enhanced execution speed with
which are classified per the computa- be in parallel in these algorithms are lower power consumption. For manu-
tional characteristics. relatively fewer and are hence not ideal facturing organizations that deploy
PROComSol
Process Communications Solutions
procomsol.com
hundreds of machine vision systems in the placement of the new hardware ABOUT THE AUTHOR
their shop floors across the world, the and routing. Once a new hardware
Gineesh Sukumaran (Gineesh.Sukuma-
savings are massive. is introduced, the time synchroni-
[email protected]) is an image processing and
● Faster time to market: As reconfigu- zation between the older hardware
computer vision specialist. He is technical
rable systems can be easily upgraded, system and newly introduced hard-
head at L&T Technology Services’ Digital
a product with reduced functionalities ware should be factored in. Other-
Image Processing division and has more
can be quickly designed and released wise, the system can behave incon-
than 14 years of experience in the design
to the market. Unlike ASIC systems, sistently.
and development of computer vision
most reconfigurable platforms have ● Development tools: The existing de-
algorithms. Sukumaran’s research inter-
simulators that developers can use to velopment tools in reconfigurable
ests include biomedical image analysis,
swiftly build prototypes and facilitate computing require a great deal of
multiple view geometry, and deep learn-
faster time to market. manual intervention. Most of these
ing. He has six patents in this field.
● Lower system cost: For ASIC and general- tools are in the development stage
purpose hardware, the design costs are and not very mature. Developers
View the online version at www.isa.org/intech/20190604.
high, and the life of the vision product need commercial off-the-shelf tools
is short. Because reconfigurable sys- to be able to handle the develop-
tems have flexibility and upgrades, the ment and implementation of recon- RESOURCES
upfront design costs are substantially figurable systems. The Current Status of Reconfigurable
lower, and the shelf life of the vision Weighing the pros and cons, recon- Computing
product can be significantly enhanced. figurable computing has immense https://doi.org/10.3929/ethz-a-004704215
promise in the field of machine vision.
Reconfigurable computing:
The drawbacks As more developers get onboard the
architectures and design methods
● Difficulty in programming reconfigu- reconfigurable wagon, the development
http://www.doc.ic.ac.uk/~wl/papers/05/
rable chips: The programming issues and implementation of these systems
iee05tjt.pdf
in reconfigurable computing involve will significantly ease. ■
atotec® celebra
om te
hr
s
Connect Your Allen-Bradley PLC to the cloud
40 YEARS OF EXPERTISE
Online Gas Analyzer Experts
in
n
ti o
n-
o
li n ta
e i n str u m e n
Move Your Data to a Web Services Application Sulfur compounds, VOCs and Odorants
You’ll find it easier than ever to make your Monitoring in Hazardous Areas
Allen-Bradley PLC accessible to your Cloud
or enterprise systems using simple HTTP
commands. Register or tag data in your PLC 3KHQRO
is exchanged with your enterprise system or
cloud application as an XML or JSON file.
www.chromatotec.com - [email protected]
I
By Jason Andersen ndustrial manufacturers understand that maintainability, among other benefits.
holding steady is rarely an option. This is Production and manufacturing industries
because ongoing productivity improve- have long employed supervisory control and
ments are a near-universal requirement. Man- data acquisition (SCADA) applications, human-
ufacturers can achieve such improvements in machine interface (HMI) software, and other
many ways, with the most valuable upgrades automation platforms, such as programmable
enhancing several things simultaneously, per- logic controllers (PLCs) and process automation
haps across multiple levels of an organization. controllers (PACs). Progressive use of these sys-
Today’s automation experts are discovering tems reduces waste, increases uptime, and im-
that moving visualization and control func- proves overall equipment effectiveness (OEE).
tionality toward the industrial “edge” is helping These systems require some level of “core”
them achieve better operations and improved server and PC computing power to function,
M
By Fawaz AlSahan, aintenance can be made more effi- ● Forty to 60 percent of PM tasks add no value
CAP, SCE cient, helping make operations more to operations.
efficient and profitable, by leveraging ● Some PM tasks are similar.
new digital technologies. Conventional time- ● Many tasks require removing the instruments
based preventive maintenance (PM) is widely or valve to the shop, while condition monitor-
used throughout industry but has shortcom- ing is enough to determine reliability.
ings that can be overcome with smart sensors ● New technologies and digital instrument
and analytics (digital transformation). This features are not used to optimize tasks.
enables process facilities to move into perfor- ● Some PM tasks have a high probability of
mance-based and shutdown-based preventive causing unnecessary shutdowns and opera-
maintenance. tion interruptions.
The common categories for preventive main- Using smart sensors and analytics (digital
tenance are: transformation) resolves these shortcomings
● Time-based preventive maintenance: Conducted and enables process facilities to move into
as a defined procedure during a preset interval. performance-based and shutdown-based
● Performance-based preventive maintenance: preventive maintenance.
Conducted when the instrument perfor-
mance goes below a certain limit. Perfor- Defining PM scope, tools, and frequency
mance monitoring is done either online (via To have the right preventive maintenance for
an online instrument) or off-line (via a por- instrumentation, first clearly define the type
table instrument). and scope of PM. Proper tools need to be se-
● Shutdown-based preventive maintenance: lected, and the right frequency defined. The
Conducted during plant shutdown. scope of PM can include a certain task to be
The diagram shows the ISO categorization for performed. This can be a time-based visual
maintenance in general and preventive mainte- check, time-based functionality check, remov-
nance in particular. al and overhauling, or a condition-based task.
It is worth highlighting that conventional Second, address who will perform the PM task,
time-based preventive maintenance is widely such as a field operator or maintenance techni-
used and has the following shortcomings: cian. Determining this point requires agreement
between the parties, with clear tasks and a sys- crepancy alarm
FAST FORWARD
tem to collect and communicate the findings. between adjacent ● Preventive maintenance for instrumentation
The tools used for PM tasks are either instru- control and safety shall be redefined as periodical, online, or
ment asset management systems (IAMSs), hand- transmitters shutdown tasks.
held tools, an instrument built-in keypad, hand ● avoiding soft-seat- ● Achieve PM optimization by enhancing the
tools, or shop activities. The frequency of a PM ed control valves design, effectively using instruments’ built-
task is based on prior knowledge from similar and instead install- in diagnostics, and utilizing new or alterna-
tive technologies.
equipment, manufacturer or supplier data, reli- ing a metal-seated
ability data, and performance prediction, and control valve next ● Digital transformation can eliminate or
greatly reduce conventional preventive
defines the task as online or shutdown activity. to a tight shutoff maintenance for instrumentation.
rotary isolation
Strategies to optimize PM valve
Optimizing the preventive maintenance for
instrumentation is needed to eliminate unnec- Diagnostics and online condition monitoring
essary activities and unnecessary cost. Exces- Available instruments are now smart and have in-
sive preventive maintenance may also cause ternal diagnostics (analytics) and digital commu-
nuisance trips or operation upset. This is an- nication. These features are effectively used to im-
other motive to optimize the PM tasks. prove the preventive maintenance program and
Implement optimization with the following eliminate or highly reduce traditional practices.
strategies: improve the design and apply tech- Diagnostic data is obtained and collected using:
nologies, capitalize on instrumentation self-diag- ● smart sensors (like smart transmitters, smart
nostics and online condition monitoring, and positioners, and smart pressure switches)
use the redundancy approach, once justifiable. ● an analytics data platform (IAMS, which re-
ceives the data from the smart instruments via
Design and technology a wired or wireless connection and generates a
Nowadays, available technologies, like smart status message with a recommendation)
sensors, bring a new dimension of reliability and ● digital connectivity, like Foundation fieldbus
minimize maintenance requirements. Capitaliz- (FF) with physical layer diagnostics
ing on the following technologies and design ap-
proach will highly reduce the PM scope and time: Redundancy
● single-rod, guided-wave radar (GWR) for level Redundancy is having a permanent or temporary
measurement reference to compare the installed instrument per-
● pressure transmitters instead of process- formance and reading to. Below are some examples:
actuated switches ● having dual circuits (or more) for axial and
● smart pressure and vibration fork switches radial vibration, bearing temperature, and
(with display and/or diagnostics) instead of fired equipment flame monitoring
conventional blind switches ● applying two-out-of-two voting, if safe and
● diaphragm/remote seal pressure transmitters practical, to avoid nuisance trips and reduce ex-
instead of tubing-based transmitters cessive maintenance
● digital vibration transmitter/switch instead ● installing control and shutdown transmitters
of mechanical
switches
1: Maintenance categories
● smart valve
positioner Before failure After failure
● single-rod GWR
and two-wire 2a: Preventive maintenance 2b: Corrective maintenance
noncontact radar
for inventory tanks
3a: Condition-based 3b: Predetermined
application maintenance maintenance
● electrochemical
gas detector for H2S
gas detection
4a: Testing and 4b: Condition 4c: Periodic 4d: Scheduled 4e: Scheduled
● infrared gas detec-
inspection monitoring replacement 4f: Immediate 4g: Deferred
test service
tor for combustible
gas detection
Figure 1. Maintenance categorization
● configuring a dis- Source: ISO 14224:2016
with the same calibration range and safety instrumented system (SIS) by re- (SIL) study/verification is conducted.
configuring discrepancy alarms for ducing the required physical preventive The SRS specifies the required testing
them maintenance. The safety requirement intervals for the SIS equipment (logic
● having permanently or temporarily specification (SRS) document should be solver, input and output devices).
installed pressure gauges to compare developed after the safety integrity level To define and further optimize the
to the nearby pressure transmitters’
readings
● local level gauges (sight glass, magnet- PFDavg
ic level indicator) or infrared cameras
to cross check the level instruments’ [ DU × TI ]
(displacer, differential pressure, radar, 2
etc.) readings PFDavg, SIF = PFDSensors + PFDLogic Solver + PFDFE + PFDPower Supplies
● checking the online temperature
sensor reading with a temperature
Safety integrity levels and performance requirements (for the entire system)
gauge, a test temperature element,
a portable temperature detector, or Probability of failure
ISA S84 safety on demand (pfd) Risk reduction factor
with an infrared detector/camera. integrity levels (SIL) Safety availability (1 – safety availability) (RRF) or (1/pfd)
3 99.9% – 99.99% 0.001 – 0.0001 1,000 – 10,000
SIS preventive maintenance
Diagnostics provided by the logic solver, 2 99% – 99.9% 0.01 – 0.001 100 – 1,000
input devices (like transmitters), and 1 90% – 99% 0.1 – 0.01 10 – 100
output devices (like emergency isolation 0 Process control – not applicable
valves) should support the overall pre-
ventive maintenance program for the Figure 2. Probability of failure on demand calculation and safety integrity level
required preventive maintenance for crew, and one is conducted by the op-
SIS devices, do the following for each eration crew. Operation tasks are visual Study/define
safety instrumented function (SIF): checks, visual inspections, or simple
preventive and
corrective maintenance
● Review the SIL study and SIL verifi- test and observation tasks. The mainte- operations
cation report or conduct a study. nance crew PM tasks include detailed
● Check the test interval (TI) for input test procedures, calibration, and physi-
and output devices recommended in cal testing that requires hand tools and Work planning
the study report. communication devices.
● Increase the TI to the maximum The objective of segregating the op-
number (such as matching the plant eration and maintenance tasks is to Work scheduling
total shutdown interval) and con- optimize the testing intervals and du-
firm if the SIL requirement is still ration. Some tasks are simple, requir- Release assignment
achieved. The preventive mainte- ing a long time to prepare the work of the work order
nance interval is based on the maxi- permit and the system that needs PM.
mum possible TI. Also, due to the internal mandatory re- Carry out the work
● For emergency isolation valves, quirements, the operations team must
if the TI for these valves (i.e., full be involved in these PM tasks. Simple
stroke test) cannot meet the plant examples are stroke testing for control Closure of the
shutdown window, introduce partial valves and emergency isolation valves work order
stroke testing (PST) with a weight of and testing gas detectors.
60 percent of the total stoke test. TI Production of reports
for the total and partial valve stroke Analysis of PM findings
shall then be clearly defined. There is great value in reviewing and
analyzing PM findings in terms of Reports analysis
Operator tasks versus technician tasks preventing failures and maintaining
There should be two different PM tasks. equipment reliability and availability. Figure 3. Maintenance work order
One is conducted by the maintenance This requirement is clearly addressed Source: EN 13460
in industry references (figure 3). optimization opportunities and life- and failures. The “big data” collected
Doing analysis requires the PM ex- cycle costs, and capturing and apply- from the instruments can be used for
ecuter to write accurate PM findings. ing lessons learned across the plant making decisions about when main-
The analysis of the compiled find- or company. tenance is actually needed. Hence, 70
ings helps identify the PM tasks that percent or more of conventional PM
add value and eliminate the “value- Impact on instrumentation tasks for both process instrumenta-
wasting” PM tasks. Also, it helps to preventive maintenance tion and equipment could be deleted.
identify repeated failures and to rec- Digital transformation will change Digital transformation will address
ommend revisiting some system de- the world of preventive maintenance the major maintenance challenges of
signs. Moreover, analyzing PM find- for instrumentation and equipment aging equipment and the aging work-
ings helps improve PM tasks by better to more “predictive analytics,” where force, and can highly cut cost, time,
managing spare parts, highlighting data is used to predict performance and manpower.
R
enee Bassett has been with Zabel to determine publication and content strategy, develop
named chief editor of ISA’s content, and manage the publications’ contributors, staff, and pro-
InTech magazine and Auto- duction processes. Bassett will be based in Nashville, Tenn.
mation.com digital content-delivery Bassett is an expert content creator, media manager, and mar-
site for automation professionals. keting communications specialist in the areas of industrial man-
Bassett replaces Bill Lydon, who will ufacturing, technology, and automation. She possesses more
continue with the publications as than 20 years of experience in content generation for techni-
contributing editor. cal and nontechnical audiences. Over the course of her career,
Bassett joins ISA’s team of pub- Bassett has provided her services to leading automation publica-
lishing professionals that includes tions—including Automation World and Control Engineering—
Lydon, ISA’s group publisher Rick and developed and implemented media and communications
Zabel, and content editor Cory Fogg. In her role, Bassett will work programs for many industrial automation companies. ■
If APC needs single-loop control to reject un- “create” an intensive variable. Ratio variables are
measured disturbances, then how best to utilize it? used quite often in process control. For example,
reboiler duty on a distillation column can be cal-
APC control objectives culated from flow and temperatures and then
The most important APC control objectives are ratioed to the column charge rate. The column
related to production rate and product qual- reboiler duty/charge ratio can then be used as an
ity, because these variables are directly related to MV in an APC application. Same for treat ratios
operating profitability. Production rate is limited in absorbers, product yields in fractionators, and
by constraints (e.g., maximum temperature and so on. The single-loop flow controller rejects un-
control valve position). Product quality is normally measured disturbances and thereby stabilizes the
controlled by temperature, analyzers, lab analyses, created intensive variable.
or indirectly by “soft sensors” (inferred properties).
Vessel levels are integrating, inventory-related vari- Status quo: Do not disturb
ables and are almost never included as CVs in an Today, even after 50 years, APC continues to rely
APC controller. The same can be said for vessel on the lowly flow control loop, the most basic
pressure, unless it is a constraint variable related single-loop control, as the best rejector of unmea-
to pushing the production rate. (There are some sured disturbances and the most stable platform
exceptions—fired heater controls often use burner for the APC/optimization control hierarchy. So, the
pressure as a substitute for fuel gas flow to control next time somebody suggests getting rid of single-
heater temperature.) loop control in an APC application, just ask, “What
The lone outlier is flow, which is a true “exten- about unmeasured disturbance variables?” Do not
sive” variable, independent of product quality or expect a righteous reply. ■
operating profitability. In all but truly exceptional
cases, flow is always adjusted to achieve some other
process control objective. It is almost always the ABOUT THE AUTHOR
secondary, or slave, in a one-on-one basic cascade. Jim Ford, PhD ([email protected]), is
If the flow controller is standalone, then its SP is a senior consultant at MAVERICK Technologies, a
adjusted by the operator (or an APC controller) to platform-independent automation solutions provider
achieve a higher-level control objective. offering industrial automation, strategic manufactur-
At the same time, the flow controller is a true ing solutions, and enterprise integration services for
mitigator of unmeasured disturbances. It is typi- the process industries. Ford specializes in advanced
cally characterized as a high-frequency loop, process control, control system engineering, and ex-
meaning that a change in OP is followed almost ecution of detailed control system modernization and
immediately by a change in PV. When tuned migration front-end loading evaluations.
properly, and when
challenged by unmea-
sured disturbances, it Error
returns the PV to its detector
SP very quickly and Actuating
signal
with little overshoot
or oscillation. As such, Error
signal
it is normally the vari-
able of first choice as
an MV for any higher-
level control strategy,
especially for an APC
Feedback elements
controller. Feedback signal
There are some ex-
ceptions. Although flow
is a true extensive vari-
able, there are instances Single-loop control (shown) is the mainstay of process control, and APC can never be successfully
where it can be used to implemented in its absence.
I
was fortunate to have the opportunity to Cappelli described the industry trend of CISOs
talk with Dawn Cappelli, an experienced and being given responsibility and/or accountability
accomplished cybersecurity expert, who shared for all cybersecurity for the company. One reason
her advice on building cybersecure manufacturing why is that cybersecurity in IT is significantly more
organizations. Cappelli is vice president, global mature than in OT, and someone with IT security
security, and chief information security officer at experience understands how to methodically build
Rockwell Automation. the cybersecurity program across the organization
Before coming to Rockwell, she was founder and using a risk-based approach.
director of Carnegie Mellon’s CERT Insider Threat One of the challenges is building a cross-function-
Center. Cappelli is recognized as one of the world’s al team including both IT and OT, since tradition-
ABOUT THE AUTHOR
leaders in insider threat mitigation and has worked ally they have not worked closely together. Cappelli
Bill Lydon (blydon@isa.
with government and industry leaders on national recommends using the NIST Cybersecurity Frame-
org) is an InTech con-
strategy issues. Cappelli is a certified information work (NIST CSF) (www.nist.gov/cyberframework)
tributing editor with
systems security professional (CISSP), and she has a as a tool to deploy a focused process and involve
more than 25 years of
BS in computer science and mathematics from the all parties. The framework helps to identify gaps in
industry experience.
University of Pittsburgh. Cappelli came to Rockwell cybersecurity strategy and becomes the blueprint
Automation in 2013 as director, insider risk, and for risk assessment. Bringing together cross-func-
“People are realizing now, due to the convergence of IT and OT, that it’s
important to have one security leader responsible for all cybersecurity
for the company.”
—Dawn Cappelli
built the company’s insider risk program. Her team is tional personnel consisting of IT and OT experts,
responsible for protecting Rockwell Automation and plant experts, and plant engineers using the NIST
its ecosystem of customers, suppliers, distributors, CSF focuses the activity and fosters team building
and partners from the ever-changing global cyber- based on shared goals. This process for building the
threat landscape. strategy creates a shared vision and understanding
I asked Cappelli what first steps a manufacturing of all stakeholders’ challenges and ongoing positive
company should take on the journey to achieve working relationships.
cybersecurity protection. She shared her experience, An important part of this process is prioritizing
insights, and recommendations for creating a com- cybersecurity efforts based on risk. This helps com-
prehensive industrial cybersecure manufacturing panies prioritize investments, because it is typically
organization. impractical to do everything at once.
The first step a manufacturer should take is to de- I asked Cappelli for any tips based on her expe-
termine the leader of the cybersecurity effort. Cappelli rience building the Rockwell Automation program
noted that many manufacturing companies already over the past few years. She suggests starting first
RESOURCES have a chief information security officer (CISO) re- in the IT group to “get your feet wet” if you have
sponsible for information technology (IT) security, but not yet used the NIST CSF, then use the NIST CSF
NIST Cybersecurity
traditionally operational technology (OT) security has Manufacturing Profile to create your manufactur-
Framework
been the responsibility of the OT engineers. “People ing security strategy. Also, the NIST CSF helps to
www.nist.gov/
are realizing now, due to the convergence of IT and identify some quick wins for the manufacturing
cyberframework
OT, that it’s important to have one security leader re- environment, like ongoing communications to
ISA Cybersecurity sponsible for all cybersecurity for the company.” This maintain security awareness among plant per-
Resources is someone who can work with both IT and OT to sonnel. Rockwell Automation has done this with
www.isa.org/technical- build and execute a holistic cybersecurity strategy that a monthly cybersecurity awareness bulletin to re-
topics/cybersecurity/ encompasses the entire ecosystem of not only IT and inforce topics like the importance of physical se-
cybersecurity-resources OT, but also of all external connections, including third curity, social engineering, not sharing passwords,
parties and the supply chain. and safely using USBs. ■
T
he rate of change for new technology in the world of project team regarding design. Again, this goes back to the
control system engineering is staggering and can be over- CSE having a strong understanding of process design and pro-
whelming if you do not step back to see the forest for the cess control.
trees. The role of a control system engineer (CSE) is changing So, the next time you start to stress about a new version of
with this technology and the cost of its implementation. Hard- an HMI or a just-released series of I/O modules, take a step
ware and software vendors continue to develop products that back and make sure you are also acquiring knowledge about
are more cost effective—from both product cost and implemen- how the equipment you are controlling actually operates. And
tation cost perspectives. how it can operate better. ■
Do not forget that a significant cost of a product is the time
required by a CSE to implement it. And let’s face it, we are well ABOUT THE AUTHOR
paid, particularly in North America and Europe. As cool as you
Michael McEnery ([email protected]) is the
might think the latest tool from your software vendor is, part of
president of McEnery Automation, a CSIA certified system in-
the reason it exists is to reduce your cost. Overall hours required
tegrator. Founded in 1994, the Control System Integrators As-
for the development phases of automation projects (program-
sociation (CSIA) is a not-for-profit professional association of
mable logic controller programming, human-machine interface
more than 500-member companies in 40 countries advancing
[HMI] programming, device configuration) have decreased drasti-
the industry of control system integration (www.controlsys.org).
cally over the past 15 years.
Technology also brings opportunities for offshore system de-
velopment, which further decreases the time CSEs are spending
with program development. At the same time, new tools provide NEW! From ISA Publishing
opportunities for CSEs to deliver even more value. Data histori-
ans; Industrial Internet of Things; data analytics; advanced pro-
cess control; proportional, integral, derivative (PID) autotuning; Situation awareness + situation assessment
and PID loop performance monitoring are all available to help + successfully managing abnormal
CSEs save energy, increase plant uptime, reduce material costs, situations = Situation Management
etc. But these tools
have no value un-
Operators must be able to
These tools have no value til they are put to
monitor operations,
work by someone
until they are put to work by who understands understand the data, and
plan and actualize neces-
someone who understands the processes be-
sary changes. This book
ing controlled. This
the processes being con- discusses the technology
places a stronger
and tools, as well as
trolled. This places a stron- emphasis on un-
effective methodologies for
derstanding pro-
ger emphasis on understand- cess control basics. safer and more productive
control room operations
ing process control basics. It It requires getting through situation
away from the
requires getting away from management.
keyboard and out
the keyboard and out onto onto the plant floor.
Control system
the plant floor. integrators also
have new oppor-
tunities for project management, design, and documentation. Order your copy today at
End users have fewer resources available for projects, and the
people who are assigned have less and less time. But a suc-
www.isa.org/situationmgmt
cessful project still requires someone to understand the process
requirements in detail and to communicate effectively to the
I
SA99, Industrial Automation and Con- is appropriate and sufficient for IIoT. This Industrial-Process Measurement, Control
trol Systems (IACS) Security, has begun assessment is vital, as the IIoT is a specific and Automation. With more than 900
work on an ISA technical report to be case of IACS with a very wide range of ob- members, ISA99 draws on the input of
titled, Applying ISA-62443 to the Indus- jects, an extended surface area, and a high cybersecurity experts across the globe in
trial Internet of Things (IIoT). Coverage will scalability, resulting in a very large attack developing the standards, which are ap-
include general categories of IIoT devices surface and new vulnerabilities. The work- plicable to all industry sectors and critical
within IACS, cybersecurity challenges, and ing group will first examine the specific infrastructure in providing a flexible and
rates of adoption in industry. The report risks and new forms of attack to which the comprehensive framework to address
will then serve as a guide in determining if IIoT may be subject. and mitigate current and future security
any new and related ISA99 work products This project is one of numerous cur- vulnerabilities in IACS.
are required. rent development activities within ISA99 ISA99 Working Group 9 is cochaired by
The technical report will be the first to support and advance the widely used Suzanne Lightman of the U.S. National
work product of Working Group 9, IIoT ISA/IEC 62443 series of standards on IACS Institute of Standards and Technology, and
and Security, which was established by security. The standards are developed Eric Braun of Emerson Automation Solu-
ISA99 to analyze the specific characteris- primarily by ISA99 as American National tions. For information on ISA99, contact
tics of the IIoT in terms of threats, attack Standards, with simultaneous review and Eliana Brazda, ISA Standards, ebrazda@isa.
surface, and vulnerabilities, and to exam- adoption by the Geneva-based Interna- org. For information on viewing or obtain-
ine whether the approach developed by tional Electrotechnical Commission (IEC) ing the ISA-62443 standards and technical
ISA99 for securing a conventional IACS through IEC partner committee TC65, reports, visit www.isa.org/findstandards. ■
I
tegral, derivative (PID) algorithm is used in industrial control systems to aid in SA75, Control Valve Standards, has
used in a vast majority of applica- their selection and application to improve completed a revision of ISA-75.05.01,
tions for basic control and in many for ad- manufacturing processes. The working Control Valve Terminology. The standard
vanced control. However, there appears group will develop technical reports, recom- provides a glossary of definitions commonly
to be a widespread lack of understand- mended practices, and standards document- used in control valve applications across
ing of the different forms, structures, ing the algorithms used in industrial control industry. The standard includes a number of
and features, and of how performance systems and the measures of performance new or significantly revised terms, including
objectives determine appropriate choices. for those algorithms. The documents may backlash, cage guiding, intelligent/smart po-
Consequently, most of the capability of include guidance on algorithm selection. sitioner, quick change trim, globe valve body,
the PID is underutilized, reducing process ISA5, the oversight committee for ISA5.9, hysteresis, port guiding, and post guiding.
safety, efficiency, and capacity, points is also responsible for the widely used stan- ISA75 is chaired by James Young of the
out ISA Fellow, author, and mentor Greg dard ISA-5.1, Instrumentation Symbols and Dow Chemical Company. For information
McMillan, a widely recognized expert on Identification. The committee is chaired on ISA75, contact Eliana Brazda, ISA Stan-
process control who received the ISA Life- by former ISA Standards & Practices De- dards, [email protected]. For information
time Achievement Award in 2010. partment vice president Tom McAvinew, on viewing or obtaining ISA-75 standards
This concern prompted McMillan to co-author of the ISA book Control System and technical reports, visit www.isa.org/
propose a new ISA standards project on Documentation: Applying Symbols and findstandards. ■
controller algorithms and performance, Identification.
which has led to the recent formation of The ISA5.9 cochairs are Yamei Chen of
Have an idea for an ISA
ISA5.9, Controller Algorithms and Per- Eli Lilly and Company, Indianapolis, and Mi- standard, book, training
formance. The new working group will chel Ruel of BBA, Québec. Those who are course, conference topic, or
function under the ISA5 committee, Doc- interested in participating in the new ISA5.9 other product or service?
umentation of Measurement and Control are asked to contact Charley Robinson, ISA Send it to [email protected]
Instruments and Systems. Standards, [email protected]. ■
Moore Industries........................................6
www.miinet.com
ProComSol, Ltd.........................................28
www.procomsol.com
Chris Nelson
Advertising Sales Representative
Phone: +1 612-508-8593
Email: [email protected]
Kelly Winberg
Advertising, Classifieds Section
Phone: +1 267-718-8197
Email: [email protected]
Chris Hayworth
Advertising Materials Coordinator
Phone: +1 919-990-9435
Email: [email protected]
Reprints
Foster Reprints will work with you to
create a customized reprint package,
including hard copy reprints, eprints,
and mobile-friendly products.
Contact Jill Kaletha at 219-878-6068
or [email protected].
D
o you have a plan to combat the fear mon- of cyber-assessment or mitigation project, and you
gers and the shiny-new-thing pushers? have not first assessed your risk profile, then you
I grew up in the plant floor automation have missed a critical step. Ask to see the data that
business. I have seen a lot of change in technology, shows this is your number one threat to operations.
and I have seen a lot of resources wasted on things I have found that in 100 percent of cases, threats
that really had no business case. Recently, this exist with a far greater impact to operations than
waste seems to be focused on cybersecurity and a cybersecurity incident. Bad business decisions are
other shiny new things. In an age where the pace made in the name of cybersecurity, like removing
of technological advancement is higher than ever, remote access from maintenance personnel, which
our profession is being bombarded daily with the significantly increases costs and response time
ABOUT THE AUTHOR
next great thing. Managers get calls from highly for troubleshooting. What are we really trying to
Dean Ford, CAP
effective sales people and do Internet searches that protect against?
dford2009@gmail.
com) is the managing
principal engineer and
If your organization is currently involved in any sort of cyber-assessment
executive vice president or mitigation project, and you have not first assessed your risk profile,
of Westin Technology
Solutions Engineering
then you have missed a critical step.
Practice. He is a licensed
teach them of all the threats they must protect I propose a different path. An overwhelming
control systems engi-
against and all the opportunities they are missing majority of cyberrisk can be eliminated through
neer in 18 states and an
because they do not have X, Y, or Z. This pressures simple procedural and policy changes that cost no
active ISA member. Ford
automation professionals to also focus on cyber- money. Instead of spending money to prevent a
participates in AWWA,
security and new technologies and usually forces cybersecurity issue, first go through the exercise of
WEF, WRF, SWAN, and
them to bypass some critical decision steps. assuming you will be hacked. How will you and the
is on the Government
Automation professionals are on the front lines of organization respond? What key decision-making
Relations and Workforce
risk management and mitigation. The cyberthreat process do you need in place? What is the hierar-
Development commit-
is merely one of many threat factors to overall risk chy? Who is authorized to call something a disas-
tee of the Automation
management, and it gets far too much attention. ter? Is your disaster recovery plan in place and has
Federation.
Our profession is getting a lot of pressure to spend it addressed things like building a new automation
money on this. “Your peers are doing it, why aren’t platform from bare metal? Have you tested it? Are
you?” your backups working? The list goes on. The key
If you have heard me speak at various conferences, point here is that if someone wants in, he or she
you have likely heard some shocking words. is going to get in, and it will be through normal
Cybersecurity is not the threat that it purports to channels. All of the money spent to prevent a hack
be. In one session recently, I was challenged. A will be useless.
prominent manager at a utility had done a lot of With a sound risk management and disaster
surveys with conclusive results. Cybersecurity was in recovery plan, you not only address cybersecu-
the top five biggest perceived threats to operations. rity incidents, but you also mitigate fires, theft,
I agree it is a threat, but I challenge the notion that weather events, rogue employees, etc. And per-
it is our greatest threat from two sides. First, if it haps, during the risk mitigation planning, cyber-
was truly that large of a threat, every utility would security jumps out as a critical need, and you have
be budgeting lots of money to address it. That is to develop some projects for network segregation.
not happening, and the more advanced utilities and At least then you will know that the capital is be-
companies are not wasting resources on it for very ing deployed in the best way and for the correct
good reasons. Second, there are far greater threats priorities. I suspect you will find that you should be
that should be dealt with first. Planning is the key to fixing the power failure scenario that causes hours
relieving the pressures. of down time, or moving the server out of the
The cyberthreat is merely one of many, many control room, or simply upgrading the automation
risks that must be addressed in our operations. If platform to current hardware with parts that you
your organization is currently involved in any sort can buy from a more reputable place than eBay. ■
© Allied Electronics & Automation, 2019 Get your A&C fill at alliedelec.com 1.800.433.5700
May/June 2019
A P U B L I C AT I O N O F T H E I N T E R N AT I O N A L S O C I E T Y O F A U T O M AT I O N
www.isa.org/intech
The Unlimited SCADA Platform
of the Future is Here