With a turnout of around 800, albeit with a considerable number of Microsoftees in attendance, the Microsoft Global Energy Forum1 has grown considerably since Oil IT Journal’s last visit back in 2009. Shell VP IT Services Jay Crotts kicked off the proceedings with a keynote on, ‘Combining IT and business skills to generate value.’ Crotts noted the power of modern consumer technology. Today, a mere graphics card costing a few hundred dollars packs the power of what used to be a multi-million dollar supercomputer. We are on the cusp of a similar evolution with cloud computing.
What does this mean for Shell with its 140K desktops, 14,500 servers, 140k phones and 10 million customer interactions per day? Shell is not ready to abandon its own servers and move to a cloud-only environment. Current thinking is along the lines of ‘hybrid cloud computing’ (HCC) that lets Shell pay for what is used, enables innovation and provides variable capacity. HCC can be carved up as follows—1) infrastructure as a service, 2) platform as a service and 3) software as a service (I/P/SaaS). Such a tiered approach lets IT ‘migrate’ from self, to supplier-managed. As an example, conventional wisdom has it that the average life of a server in Shell is around 2 years. But HCC dispenses with this notion. In Houston, servers can be optimized in 4 hours.
Security is still perceived as a big issue—even though this is maybe not so bad with a reliable supplier. For Crotts the key is how to extend Shell’s security model to the cloud. Global regulation and export compliance may restrict what can run outside of a given jurisdiction.
The HCC has allowed Shell to ‘seamlessly’ integrate with Microsoft’s own solutions including SharePoint 2010, Windows Azure—offering IAS, SAS and PAS. Shell has partnerships with Microsoft on SharePoint, Azure and telecoms. Crott believes that Microsoft is leading the world in the breakthrough technology of the cloud.
Craig Hodges led a Microsoft team demonstrating the ‘Contoso Oil’ proof of concept scenario. Here real time tracking of a compressor spots vibration trouble and initiates a timely intervention from the field engineers. The PoC included a bewildering number of Microsoft products—StreamInsight, Dynamic CRM, Office 365, HyperV and Windows 7. OSIsoft RT web parts and ‘Project DaVinci’ also ran. An intriguing new use of customer relationship management (CRM) involved extending the definition of ‘customer’ to include compressors and pumps! And of course, heavy duty number crunching was performed in the Azure cloud.
Ken Startz recalled that
Marathon’s first OSIsoft PI System was installed in a refinery in 1988.
The company now has four in upstream, seven in downstream, and another for supply
and distribution. The systems are now a lot easier to deploy and manage on today’s
Microsoft platform than back in the days of VMS. The systems integrate with
Marathon’s ‘ViewPoint’ digital oilfield for data mashups through the SharePoint
standard. This supports integration with other commercial software from Kappa
Engineering, Halliburton, AspenTech and Schlumberger—blending data from six
different control systems. Knowledge management is simplified as new hires
only have to learn one technology. The ubiquitous RT Webparts are used by engineers,
but also in HR and HSE. Marathon’s flagship Alba field in Equatorial Guinea
is now supported by a single PI System, with all data mirrored back to Plano.
The huge GE compressors are monitored remotely and debottlenecking is performed
remotely. Marathon puts the value of its digital oilfield effort at around
0.1% of its worldwide production. The system has enthusiastic backing from Marathon’s
executive VP upstream Dave Roberts. Startz concluded, ‘If you want to be successful,
use Microsoft and OSIsoft tools.’
CIO Abraham Galan described how Pemex
is hoping to leverage IT to help reverse the significant decline in output it
has seen in the last decade during which Mexico’s production has decreased from
3.3 to 2.6 million barrels per day. Pemex is structured as four separate companies
with great autonomy and little top-level governance. Data is scattered all over
the place and it is a challenge to realize the value—particularly in the context
of the constraints imposed by Pemex’ cost reduction initiative. IT investment
optimization has led to the adoption of a common platform (SharePoint), common
standards—PPDM and ‘MURA’ and lower infrastructure costs through use of the
cloud. Pemex’ deployment focuses on applications that leverage all of the above,
notably iStore’s ‘PetroTrek,’ used to create a single source of the truth across
multiple data sources. Data is the key to unlocking the value in Pemex’ assets,
‘If people can’t find data they will either make it up or rework it!’ Galan
noted the knowledge gap between different stakeholders. Data gatherers, interpreters,
users and curators each have their own, different versions of the truth. Data
sharing is a ‘challenge,’ ‘it is not part of our culture.’ Data can be isolated
and invisible. Both centralized and local data stores have their own problems.
Galan advocates a middle road—building a single source of the truth by accessing
the closest original source—although this is often hampered by data formats,
tools and ‘irreconcilable’ technologies.
Data will be key to Pemex’ new incentive-based mature fields program. This heralds a new era for Mexico with new players invited in and represents a low risk opportunity for foreign companies to get revenue by applying new technologies. Pemex’ National Data Repository leverages Microsoft SharePoint with encouraging results. In one field, certified reserves have doubled. Data-driven buisness intelligence helps discover new fields and increase recovery from existing ones. Pemex is very pleased with iStore’s know-how and technology. In the Q&A Galan was asked what SharePoint has brought to the table. He noted that building a data repository required a lot of configuration to access existing data sources. This is easy in SharePoint. The decision was also cost driven—a Microsoft frame agreement means that new SharePoint instances are essentially free.
Eric Rearwin outlined Chevron’s work with Invensys/IntelaTrac on a mobile system to support its operator routine duties (ORD) compliance and monitoring effort. This monitors operator data collection, maintenance and greasing activity at Chevron’s 22,000 retail outlets across 6 continents. The activity was previously monitored with paper run sheets and logs. But a lot of exception-based information was either un-recorded or ‘stranded’ and unavailable to other stakeholders. Half the assets in a refinery are un-instrumented—and clipboard data is not timely. There is great potential for rotating equipment monitoring and condition-based maintenance.
Chevron’s solution is to plug vibration monitors into IntelaTrac and to deploy a standard electronic mobile system for ORD. This feeds systems of record in laboratories and data historians. Handheld-based data collection includes ‘proactive’ data analysis, higher equipment availability and faster reaction to changing plant conditions. When Katrina struck, email instructions and alerts meant that operators from other plants could be rushed into action. Tank monitoring makes operators aware of heavy rain—avoiding tank collapse. Other applications have been deployed in the rotating equipment arena. These are ‘$10,000 fixes’ as opposed to $100,000 problems. IntelaTrac is now working its way upstream into Chevron’s E&P units and the ‘field of the future.’ All runs on a Microsoft stack—including Windows Mobile. While technology is key, behavior also needs consideration. It is easier to puzzle over how much memory to load into a hand held device than to check that operators are behaving appropriately. Sites still operate independently—sharing best practices is challenging—especially in the upstream. It is also key that data is actionable—a big complaint is that collected data is not acted on.
Evan Bauman from Shell’s Cri-Criterion catalyst unit likened the way Shell monitors its catalyst performance to how GE monitors its rotating equipment. But you can’t put a sensor into a 1/16th inch pellet, so catalyst monitoring is done by observing process variables. The value of a catalyst is largely determined by how customers use them. Shell has hundred of users across multiple time zones. Data was previously stored on local drives and collaboration was based on email. Stuff was easily lost in .pst files and there were disconnects between customers and catalyst information. Then Shell deployed its IT ‘standards’ (i.e. SharePoint), first in a 2009 pilot with Logica which built the CatCheck Portal. CatCheck plugs into SAP and offers a discussion board, search, customer zone—all linked with business process. While there have been observable business benefits, it is ‘hard to put a number on them...’
Also new at the GEF was Idea Integration’s ‘Constellation,’ a SharePoint/ESRI ArcGIS Server combo that targets the upstream. Constellation GIS includes integration with geoscience desktop applications through an OpenSpirit data link.
More from David Barret, dbarret@microsoft.com.
In 2009, we described SharePoint as the centerpiece of the GEF. This time we were expecting to hear more of MURA2. This did not happen. The GEF remains essentially a group meet of very enthusiastic oil country SharePoint users. Maybe MURA will get a session or two in 2012—if anyone can figure out exactly what it is!
1 Download presentations from www.oilit.com/links/1102_3.
2 Microsoft Upstream Reference Architecture.
© Oil IT Journal - all rights reserved.