APRIL 26, 2016

ExxonMobil’s Quest for the Future

of Process Automation

By Harry Forbes

Keywords

Cloud Computing, DCN, DCS, DDS, ExxonMobil, FACE, Future Airborne

Capability Environment, Open Source, Process Automation

Overview

This ARC Insight is taken from an address to the combined Business Secu-
rity, Information and Communication groups of the International Business
Congress (IBC) on April 8, 2016 by ARC Research Director Harry Forbes.

ExxonMobil’s recent process automation initiative is noteworthy for several

reasons: First because it is being done for ExxonMobil, a leading interna-
tional oil company with a long reputation for operational excellence.
Second, because the products of this program will
ExxonMobil has shaken up the process be technologically quite different from the process
automation market by beginning a automation systems used today. Third, because the
program to develop both a new process
organization and the value chain executing this
automation system architecture and a
program are also quite different from the way the
new process automation value chain.
This report is taken from a recent process automation market works today.
address by ARC Research Director Harry
Forbes discussing the origin and scope ExxonMobil discussed the objectives of this initia-
of ExxonMobil’s program. tive at the 2016 ARC Industry Forum in Orlando in
February, where it generated significant discus-
sions among technology end users and suppliers alike. The following ma-
terial represents a combination of publically available information and
ARC’s own assessments.

During their 2016 ARC Industry Forum presentations, ExxonMobil repre-

sentatives stated clearly that one objective in making their automation plans
public is to promote the development of a business “ecosystem” (a market
segment with many companies supplying products and services). This eco-
system eventually would support and enhance the new automation
technology that ExxonMobil is adopting. They explicitly said, “We do NOT

VISION, EXPERIENCE, ANSWERS FOR INDUSTRY

 ARC Insights, Page 2

want to create a system that is only used by ExxonMobil. That would rep-
resent a failure of this program.”

Strategies are being developed now with the objective to enable multiple
end user and supplier firms to effectively collaborate on the standards and
practices that will be required for the ExxonMobil automation development
program. These plans should be announced shortly, probably in the second
quarter of 2016.

Key Differentiators of ExxonMobil’s Program

The plants for which these automation systems are being developed are
those of the downstream and chemical operations of ExxonMobil. This
program is noteworthy for many reasons. In ARC’s assessment, the three
most important distinctions are:

• First, the organization of the program and the value chain is different
from the way today’s process automation market is structured. The
system integrator has a different role. The firm that serves as the sys-
tem integrator does not supply any of its own hardware or software to
the program.
• Second, the effort will adopt a rigorous software architecture that is de-
signed to optimize portability of software and to make the resulting
systems highly interoperable, easily extensible, and more modular
compared to today’s automation systems.
• Third, the program will include new electronic equipment which is
dedicated to the management of a single control loop (for example con-
trol of a single liquid flow or tank level). This represents a return to a
practice from several decades ago, and a departure from the practice of
today’s distributed control systems (or DCS) in process automation.

ExxonMobil Automation Context

ExxonMobil’s downstream and chemical operations represent a very large
installed base of plant and automation systems. There are dozens of pro-
duction sites, hundreds of process unit involved, and literally hundreds
and hundreds of large automation systems. And most of these systems are
very old, especially certain parts of them. Much of the regulatory automa-
tion equipment dates from the early 1980s

This equipment has been carefully maintained for many years, but issues of
available parts and also of support for such old technology are becoming

©2016 • ARC • 3 Allied Drive • Dedham, MA 02026 USA • 781-471-1000 • ARCweb.com

 ARC Insights, Page 3

more pressing. It was clear that despite the efforts to prolong their life, the
use of these systems would have to come to an end in the next 10 years or
so and they would require replacement. Internally, ExxonMobil sought a
way to realize some incremental business value from replacing these sys-
tems. In the company’s evaluation, just replacing these old systems with
today’s equivalent systems would not provide this value (apart from longer
expected life), and so they looked for a different kind of replacement solu-
tion. This requires a little further explanation.

Within ExxonMobil, as with most process manufacturing companies, the

automation and production management systems are thought of as hierar-
chical layers or levels. Each component operates at a specific level. The
basic regulatory controls, the base-level automation of the process is desig-
nated as “level 1.” These might control a single process temperature, flow,
or level. The tasks of higher level functions are things such as providing
broader situational awareness and driving the operating point of the over-
all process to goals that meet business-related objectives. It is at these
higher levels that the economic optimization of production operations oc-
curs.

ExxonMobil Automation Level Designations

(Sources: ExxonMobil, ARC)

Historically the ExxonMobil automation investment strategy has been to

focus its investment in these higher levels, where return on investment was
higher. The level 1 equipment is well maintained, but has not received
nearly the same level of ongoing investment. It is the level 1 equipment

©2016 • ARC • 3 Allied Drive • Dedham, MA 02026 USA • 781-471-1000 • ARCweb.com

 ARC Insights, Page 4

that is now the target of replacement, and there is a great deal of this
equipment to replace, as previously stated.

However at the end of its operating life, refreshing or replacing the level 1
equipment (DCS controllers and DCS I/O equipment) is technologically
complex and very disruptive to operations. This equipment is closely cou-
pled to both the field and to higher automation levels. Replacement
requires system-level projects that cannot easily be subdivided into very
small steps. ExxonMobil wanted very much that their new replacement
level 1 systems would never again require such a complex and disruptive
replacement program. The company wanted new automation that was
substantially simpler to refresh. So they began to envision replacing their
level 1 equipment with a very different type of system.

Developing an Automation Vision

The company developed a detailed vision and strategy for replacing these
systems. This was done internally using a small group of experienced
thought leaders from within the company who
were temporarily removed from their normal
work responsibilities for a period of several
months and given the task of precisely defining
this automation vision. They collaborated to
develop several descriptive documents.

This was all done through purely internal

work, but in February 2015 ExxonMobil dis-
cussed this vison publicly (though in general
terms) at that year’s ARC Industry Forum in
Orlando. Following that event, the company
continued to discuss the vision privately and in
A Sequestered Team Developed and
Documented a Future Automation Vision more detail with a number of automation and
system suppliers and potential suppliers. In late
2015, ExxonMobil announced a contract with Lockheed Martin. This con-
tract calls for the development of a new set of process automation products.

Components of the Automation Vision

The chart below shows the envisioned architecture. The new components
are colored yellow, while the existing systems are light blue. The new capa-
bilities are broken into three major areas:

©2016 • ARC • 3 Allied Drive • Dedham, MA 02026 USA • 781-471-1000 • ARCweb.com

 ARC Insights, Page 5

• A new Operations Platform. We have all heard of IT and OT. This is a

new type of OT platform that will be implemented using highly stand-
ardized IT-like software and hardware. ARC believes that it’s likely
that this will be implemented using an on-premise cloud platform with
some additional real-time capabilities and will make extensive use of
virtualization and of open source software.

• Second is a real-time data services “bus.” This is this set of data services
will tie the system together and enable incremental expansion and
change. This also may be implemented using open source software, but
regardless the definitions of the services will certainly be public and
probably already standardized.

New Components of ExxonMobil Automation Vision

(Sources: ExxonMobil, Lockheed Martin)

• Third is a dedicated single loop controller module, given the acronym

DCN – for distributed control node. This is a highly distributed edge
module and there may be a great many in each system. In most cases, a
DCN will regulate just a single control loop. This has the advantage of
limiting the span of an automation control module failure. Today’s
DCS controllers may manage hundreds of loops, making them much
more critical system components. Over time, existing DCS functions
migrate either to the DCNs or to the real-time operations platform.
Both of these will have their own computation and storage resources,
albeit of different orders of magnitude.

©2016 • ARC • 3 Allied Drive • Dedham, MA 02026 USA • 781-471-1000 • ARCweb.com

 ARC Insights, Page 6

A Software Architecture for Interoperability and Re-use

Besides the system architecture, software is an absolutely critical compo-
nent of the program deliverables; truly the most important part. Therefore,
the program will adopt a rigorous software design framework. This
framework serves to isolate platform dependencies and require that soft-
ware components use only fully specified interfaces for their interoperation.

What is notable about this frame-

work (named the Future Airborne
Capability Environment, or FACE)
is that it has been developed over
the past several years by firms
serving the U S Department of De-
fense. The framework is now used
to develop and procure many avi-
onics systems for US military
aircraft, including unmanned air-
craft, and its use within the
Department of Defense may ex-
pand further in the future. The
overall reasons for using FACE in
defense projects is to speed up sys-
tem delivery through re-use of
validated software components
and to enable rapid and simpler
incremental improvements to avi-
FACE Software Architecture Segments and Interfaces
onics systems installed in fleets of
(Source: FACE Consortium, The Open Group)
aircraft.

Please note that these objectives of FACE in military avionics programs

(rapid development and incremental expansion/improvement) align very
well with the objectives that ExxonMobil articulated for the replacement of
its level 1 or regulatory automation systems.

ARC Assessment
While it’s not possible to make any definitive conclusions at this early stage,
ARC’s preliminary assessment of this program covers the risks, ecosystem,
schedule, and potential benefits.

©2016 • ARC • 3 Allied Drive • Dedham, MA 02026 USA • 781-471-1000 • ARCweb.com

 ARC Insights, Page 7

The greatest risks from the perspective of end users is the need to integrate
a number of software capabilities for configuration, operation, monitoring,
and management of such an automation system. Some key features of DCS
software have historically been difficult to implement. Examples are online
control parameter and configuration changes, alarm management, system-
wide I/O and data use analysis, and online device and system software
backup and upgrade. This type of software needs to be rock-solid through
years of continuous operation, since it fundamentally manages the process
operations. Any new technology program cannot compromise the opera-
tional integrity of the plants.

Second, it will be necessary for the program to cultivate a support ecosys-

tem of businesses. While ExxonMobil envisions that in this new
architecture many routine maintenance tasks should become much simpler,
the lifetime of process automation systems is measured in decades, making
it critical to have ongoing support channels through the system life cycle.

Finally, the schedule for this program is very ambitious, with the first actu-
al installations now scheduled for 2019.

The potential benefits are those envisioned by ExxonMobil; a simplified

automation system architecture, ease in updating or extending the system
and thus smaller and more incremental system upgrades and extensions.
Another benefit may be greatly reduced dependence on any single automa-
tion supplier going forward, albeit that the overall system and its
architecture will inevitably require life-long support.

While this program carries considerable risk, ExxonMobil’s evaluation is

that now is an appropriate time, technologically, for the process automation
industry to move in an important way toward a new architecture featuring
much higher levels of system interoperability and much smaller modularity
than is common in DCS products today.

ExxonMobil may well be correct in that evaluation.

For further information or to provide feedback on this Insight, please contact your
account manager or the author at [email protected]. ARC Insights are pub-
lished and copyrighted by ARC Advisory Group. The information is proprietary to
ARC and no part may be reproduced without prior permission from ARC.

©2016 • ARC • 3 Allied Drive • Dedham, MA 02026 USA • 781-471-1000 • ARCweb.com