Scribd
Upload
78 views6 pages

Shell Leverages Data To Transform From Reactive To Predictive Operations

shell & PI

Uploaded by

Mickey Sabu
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
78 views6 pages

Shell Leverages Data To Transform From Reactive To Predictive Operations

shell & PI

Uploaded by

Mickey Sabu
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 6

DECEMBER, 2016

Shell Leverages Data to Transform from


Reactive to Predictive Operations

By Amruta Kanagali and Valentijn de Leeuw

Keywords

Business Analytics, Data model, Model-based Analytics, OSIsoft, Predictive


Operations, Shell

Overview
ARC Advisory Group analysts attended the OSIsoft EMEA Users Confer-
ence in Berlin, Germany in September 2016.

John de Koning, Technology Manager, Foundation Services, Shell Global


Solutions, spoke at the conference about the company’s
OSIsoft recently held the EMEA
goal to transform its enterprise operations from reactive to
Users Conference at Berlin,
Germany. This insight covers predictive, by treating data as an asset and deriving max-
Shell’s experience in successfully imum value from it. For this massive project, Shell
working alongside OSIsoft for leveraged OSIsoft’s PI Collective and Asset Framework
improving its digital capabilities. (AF) to collect and structure data for further insightful
Shell aims to transform its analysis. The company also carried out a proof of concept
operations from reactive to
project (POC) of applying business analytics to operation-
predictive by recognizing data as
al data on a use case of carbon capture and storage at its
an asset and deriving maximum
value from it. facility in Canada, for which it utilized OSIsoft’s PI Inte-
grator for Business Analytics for integrating PI System
data with Microsoft Power BI.

The Business Case for Predictive Operations


Shell is an innovation-driven global conglomerate, and invests more than $
1.0 billion globally in R&D each year. One aspect is to improve asset per-
formance by leveraging the huge volume of data generated in plants and
achieve a gradual transition from a reactive to a ‘predictive’ organization.

“The business needs for operations or maintenance are where and how to
quickly access information about the operation”, Mr. de Koning explained.

VISION, EXPERIENCE, ANSWERS FOR INDUSTRY


ARC Insights, Page 2

“Is the operation on target? How does it compare to similar operations?


What is the health status of the equipment? How can downtime be pre-
vented?” These are questions that need to be answered quickly and
efficiently at the local and enterprise levels.

In the past, Shell answered these types of questions by implementing is-


lands of systems, including several brands of historians for data collection.
Other software applications such as ERP or LIMS were added and used on
a local basis and based on these capabilities, local Shell staff included fur-
ther functionalities and analyses to do their tasks. But with the growth of
the enterprise and the interest in data, a need for interfacing between all
these local systems arose that turned out to be very complex. Realizing that
having each plant designing and im-
plementing systems, and that integrat-
ing the existing patchwork of systems
was not effective, the company
changed its approach about five years
ago and devised strategic responses to
data usage.

Shell recognized that along with phys-


ical assets, data is a vital asset and if
Source: Shell managed properly, it would help in
achieving its plant and enterprise-wide goals. According to Mr. de Koning,
it’s important for personnel to easily access local and global data across the
enterprise, without reinventing the wheel each time. Efficiency in retriev-
ing data and performing analyses is critical in times of depressed oil prices.
Reducing latency in taking action can help in preventing unplanned
equipment downtime, and improving asset performance. Comparing
equipment across plants, gives opportunities for improving the overall
plant performance. The power of predictive analytics would further add to
the effectiveness of responses.

The Foundation: Data Integration and Access


The first step in the series of strategic responses by Shell was to collect op-
erational data and make it available globally. The company standardized
on the OSIsoft PI System and partnered with OSIsoft for real time data col-
lection. Shell created PI System Super Collectives. These are PI Servers
holding all operational data for a line a business globally. ARC’s under-

©2017 • ARC • 3 Allied Drive • Dedham, MA 02026 USA • 781-471-1000 • ARCweb.com


ARC Insights, Page 3

standing is that these are replicated from the local PI Servers. These collec-
tives currently process close to 7 million tags together, with an
Asset Framework
update frequency of 1 to 60 seconds.
AF normalizes tag and
asset naming, units of To dramatically improve the ease of interpretation and access, a
measure, and times
uniform data model was created and implemented in Asset
zones. This is
Framework (AF). For each type of equipment, a standard data
fundamental to the ability
to standardize and enable structure was defined, capable of modeling all possible instances
collaboration across of similar equipment type. With this in place, it becomes
people and systems. straightforward to compare equipment performance worldwide,
because local tags are now organized per a global uniform stand-
ard, and have gained meaning for all users in the organization. Data mod-
elling is very crucial as a basis for smart solutions, said Mr. de Koning. “If
the same asset definition and data modeling is used for equipment
throughout the organization, then the benefits are immense,” he men-
tioned. Creating the data model took Shell several years by several teams.
Mr. de Koning recommended designing this step carefully, and executing it
systematically and thoroughly, or else one would be inviting a serious risk
of failure.

Shell’s Architecture for Data Integration and Access

Other components found in the “Data Integration and Access” layer are
event detection and advanced calculations, also using OSIsoft’s AF, and
alarm management using Honeywell’s DynAMo. Finally, OSIsoft PI

©2017 • ARC • 3 Allied Drive • Dedham, MA 02026 USA • 781-471-1000 • ARCweb.com


ARC Insights, Page 4

Coresight and PI ProcessBook plus Telerik’s KendoUI are used for data
visualization.

Mr. de Koning commented that putting all data types in a data collective
for time-series is not effective, which is why the data layer connects with
SAP HANA for relational data. This allows combining efficiently real-time
information in the context of the assets, with events and
Data Modeling and Asset
maintenance records from SAP, as meaningful input to
Models
predictive analytics solutions.
Data modeling in the sense of
Shell’s presentation is creating a
Shell also uses AF to reference data in stores other than PI
generic data structure for each
Systems, for example for static data such as limits, found
type of equipment that will fit
each physical occurrence of the
in separate ‘limit stores’ and LIMS. Shell’s ODATA que-
equipment type. ry tool can access the layer and find data across these
This needs to be distinguished sources based on the standard data model structure in
from a model of an equipment AF.
instance, a set of values in the
data structure. These objects can Shell leveraged as far as possible off-the shelf solutions
be smart for example by and software to reduce development and maintenance
containing analytics.
costs. This entailed integration of the software compo-
nents. In the case of the OSIsoft PI System, integration
required a small customization. Mr. de Koning stressed that strong partner-
ships with suppliers, and their alignment with Shell’s vision and expected
outcomes are the key success factors in implementing a roadmap.

Smart Applications for Business Value


The company built portals with advanced capabilities such as proactive
monitoring, exception-based surveillance for reservoirs, and condition
based maintenance monitoring. When adding complex calculations with
MATLAB, for example to monitor rotating equipment or fouling of heat
exchangers, Shell found that creating one calculation with input from a
standardized data structure, enabled the calculation for thousands of
equipment of this type in many facilities. “The work load was reduced
from years to days”, Mr. de Koning said.

Shell currently adds complex model-based analytics using MATLAB and R


programming languages, taking the first step in ‘predictive operations’ and
aims to make use of those tools for a new, asset-oriented way of working.

©2017 • ARC • 3 Allied Drive • Dedham, MA 02026 USA • 781-471-1000 • ARCweb.com


ARC Insights, Page 5

From Model Based Analytics to Predictive Operations


The most recent step Shell took in their transition towards predictive opera-
tion is to test the use of Business Intelligence tools, to provide ‘free format
data analytics and reporting’ as a self-service for a broad range of users.
Shell wanted this reporting to be easy to use, including the data cleansing
and preparation, the actual analyses and sharing of the results.

For testing predictive operations, Shell undertook a POC applied to carbon


capture and storage at the Albian oil sands bitumen production in Canada.
When upgrading by hydrogenation, CO2 is emitted as a byproduct of the
hydrogen production. To reduce contribution to global warming, the cap-
tured and compressed CO2 is injected and stored in a geological formation
about 2300 meters underground. For regulatory and environmental rea-
sons, field instruments monitor
the gas concentration at the
wellhead. MATLAB calcula-
tions estimate from measured
concentrations and weather in-
formation whether CO2
concentrations remain within
allowed limits.

Shell installed OSIsoft’s PI Inte-


grator for Business Analytics to
Source: Shell transfer selected data from the
PI Supercollective to Microsoft’s Power BI analytical tool. The company
made, amongst others, graphs of CO2 concentrations over time, which
compared the results from different instruments and compared concentra-
tions day by day.

Shell found the installation of the Integrator took less than four hours; it
could build reports within another four hours and considered this very fast.
The results could easily be displayed in Power BI, and easily shared in
HTML5 format on any portal. Shell appreciated the flexibility of setting
refresh rates and selecting data using filters, without any coding. As the
test finished ahead of schedule the company implemented a supply chain
use case with the same success.

Shell considers the POC to bring data analytics to a broad audience as suc-
cessful. The company aims to use the application for stock management

©2017 • ARC • 3 Allied Drive • Dedham, MA 02026 USA • 781-471-1000 • ARCweb.com


ARC Insights, Page 6

and logistics, as well as margin visualization. Shell will conduct further


tests with other Business Analysis tools such as Element Analytics

Conclusion
In the case of Shell, the strategy of transforming a reactive mode of opera-
tion to a predictive mode of operation aims in the first place to reduce
downtime, and dramatically increase the efficiency in retrieving and ana-
lyzing operational and maintenance data. Mr. de Koning’s presentation
indicated that despite the clear vision, the journey was hard work, “Shell
has examined and sifted through numerous ways of achieving the desired
results, and selected the best option that worked for it.” This also implies
that other companies may need to find their own optimal solution, adapted
to their own strategies.

Recommendations
Operational and maintenance data is an enabler for organizations to re-
spond proactively to both internal issues and external challenges. To
leverage this asset in a global organization in an efficient way, companies
must:

 define a digital roadmap consistent with the company strategy, which


includes a solution architecture that can accommodate for a “develop
once, deploy global” approach to smart add-on solutions;
 preferentially use common off-the-shelf software;
 make data globally accessible and interpretable based upon a uniform
asset-based data model; and
 leverage data in smart applications for performance analytics (report-
ing), predictive analytics and prescriptive analytics to operate with
foresight and deliver business benefits.

This article was written by ARC Advisory Group based on a presentation


we witnessed at the OSIsoft Users Conference for EMEA, and may not
completely reflect Shell’s opinion.

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

©2017 • ARC • 3 Allied Drive • Dedham, MA 02026 USA • 781-471-1000 • ARCweb.com

You might also like