Over 500 turned up for the 2001 European GeoQuest “Connect” Forum. Schlumberger is in the process of re-engineering both its software and organization. In the past we had to wonder whether GeoQuest was a subsidiary or a brand name and to cope with a matrix organization inter-fingering with ‘geo-markets’ and operating units.
We now have to get used to some more organizational concepts, such as the I-Reservoir unit, and the fact that Schlumberger is to become an I-Solutions provider, when, that is, Schlumberger is not an E-Solutions Provider! For the official explanation of the newly formed Schlumberger Information Systems (SIS) unit see page 8 of this issue.
SIS president Pai identified two problems facing the upstream, that the return on investment is too small to interest investors, and that the industry workforce is aging, with an average of around 45. Companies are faced with a bewildering choice of technology and service providers, how do you distinguish between what’s real and what’s not?
Schlumberger’s offering is rooted in its traditional technology business and is growing strongly. Others, notably the dot coms, have poor foundations and are now suffering the consequences. Pai invited the audience to become partners in Schlumberger’s e-transformation, “Oils, service companies and governments need to collaborate to get into the new economy.”
For BP’s Craig Smalley, “It’s hard to get excited about data management, but if you solve the problems, you can do exciting things - like creating space for innovation.” The objective is to improve geotechnical productivity and to be able to cope with new data types such as 4D seismics. More generally, BP wants to “move work to people, not move people to the work.” In addition, Smalley forecasts that in two to three years time, BP will be doing twice as much work with the same staff, hence the need for improved productivity and to “work smarter.”
An internal analysis showed that BP employees still spend 40% of their time on data management - which is too much. BP are addressing these issues with the Global Information Management (GIM) program outsourced to GeoQuest. GIM objectives are to let getoechnical staff focus on their core business of interpretation, to create tools, procedures and services to deliver quality analysis and information, to ‘be global’ and to fulfill the expectations of business owners while providing connectivity with vendors, consultants and other third parties. GIM aims beyond the delivery of data to geotechnical staff, to ‘help them work better.’
I know now!
GIM is to be implemented in two phases across 5 sites in Europe and N. America. GIM phase one includes the BP GIM Portal (a.k.a. iknownow!). Internal studies identified 47 functional requirements for the portal such as archival (along with context), standard report formats, reformatting capability, graphical data viewers, interactive maps of area of interest and tracking request progress. These have been wrapped up into the ‘iknownow’ straw man, incorporating Badley’s Open Journal. For Smalley, ‘business improvement’ is the objective, but this requires “deep and continuous business involvement” in product development. Smalley believes that post-implementation follow up is key to such a program’s success “We hope that this will not be seen as a ‘project’ but as the ‘currency of the day.’”
ArcView on GeoFrame
Louwaars, from Shell/Exxon’s NAM unit described how Landmark had previously developed an ArcView front-end on top of its OpenWorks database. Last year (2000), NAM adopted GeoFrame as the corporate wide subsurface integration platform. To reproduce the same ArcView functionality, NAM had GeoQuest build an ArcView to GeoFrame link which is used to visualize subsurface and surface features simultaneously. Louwaars claims the development is a powerful visualization aid.
A demo begins with a world map and zooms in to show well spider plots and 3D seismic extents. Spatial queries such as “show wells within 10,000 feet of this pipeline” can be launched and GeoFrame is accessible from new ArcView icons, using the GeoFrame ‘Open Door’ facility. ArcView on GeoFrame includes utilities to transfer GIS and oilfield data from GeoFrame to ArcView, to export a selection of ArcView objects back to GeoFrame and to create an ArcView theme from dynamically selected data in GeoFrame. Integration between the products is tight thanks to Avenue scripting plus code from the GeoFrame Development Kit. Louwaars’ underlying philosophy is that you should “do GIS in the GIS.” Schlumberger will likely roll this development into a future release of GeoFrame.
The heightened atmosphere during the presentations of GeoFrame 4 was only partly due to the extremely small auditorium - creating large spill-over crowds straining to hear what was happening. GeoFrame 4 “lets you work the way you want to work,” rather than being forced through a series of unnatural actions. One process - building a cross section in StratLog - has been reduced from ten down to two operations. Interpretations are now stored in the GeoFrame database and base mapping is now a common functionality.
GeoQuest wants us to think of GeoFrame as the upstream equivalent of Microsoft Office. The software is said to ‘breakdown the barriers between domains’ - for instance letting a geophysicist commence structural modeling and hand over to a co-worker. ‘Everyone can be involved in modeling.’ Another novelty is the integration of SeisClass, GeoQuest’s StratiMagic look-alike, into the GeoFrame workflow. SeisClass offers automated or guided, neural net-based seismic classification.
Another novelty is the ‘geo-object’ which lets the interpreter locate and map a geological unit on the seismics, generate a ‘geo-body’ and place it, as an integral element, into the geologic model. Geo-objects can include channels, levees, sheet sands etc. and may have meta data associated with them. Geo-objects can have pre-determined internal geometry (toplap, downlap etc.), and external truncations and pinch outs can be represented. Once a geo-object is categorized and validated, it can be stored in a ‘geo-analog’ database. This is a reference database where maps, images and interpretations can be grouped together and recorded for posterity. Subsequently, interpreters will trawl the database looking for examples of analogs for future work. A new ‘automated’ age-based structural framework pervades GeoFrame 4. This introduces a geological formation lexicon which can be configured to local taxonomy and stratigraphy.
Santos’ PARS project involves the replacement of legacy production accounting systems for its Queensland and S. Australia operations. Finder’s production extensions were customized to supply produced volumes, masses and energy into to Oracle Financials. The innovation was the back-allocation of produced volumes to individual wells, with estimation of produced volumes at the wellhead from sparse flow meter measurements. Two products that will likely be incorporated into future Finder releases have emerged from this work; back allocation and a production network visualization tool. GeoFrame 4 is scheduled for release in June 2001.
According to Shell’s Kramer, “Shell is not like Exxon or BP, no single person says ‘thou shalt do decision analysis.’” Such tactics are delegated to the operating companies. But when the oil price took a plunge a couple of years ago, Shell’s parent company realized the interrelation and lack of robustness of its world-wide portfolio. Portfolio analysis is now part of the E&P planning cycle and capital allocation includes global, top-down setting of ceilings “so that we can pay our shareholders.”
Economics, volumes and five year reserve data used to be stored in some 7,500 Excel spreadsheets - with an attendant 500 Power Points! Shell checked out commercial portfolio management software and found it lacking. So they have worked with Merak to extend Peep into Shell-specific products CapIT and ePlan. These deploy Shell’s standard software tool, Business Objects. Notwithstanding the new software, Shell is still struggling with the question “Is it tools or ideas that call the top-level shots?” Meanwhile, Shell is seeking to develop a standard (XML?) template to collect top level financial and other raw data from its subs throughout the world.
Norske Shell has successfully deployed both GeoQuest and Landmark applications on top of its openWorks database via the Open Spirit middleware. Pre-release versions of VarianceCube and CubeMath (in Landmark environment) have been tested and the results show the system is very stable and performance is “good, sometimes even better than existing software.” Meanwhile, Shell’s EP Technology Application and Research unit is using Open Spirit to connect its developments to industry data stores. At present, these applications ‘deal with a rather limited data footprint.’ Shell is working on extending the Open Spirit footprint to support structural modeling and engineering applications. When operational, these will be offered to Open Spirit Corp. for inclusion in a future release of the commercial product.
GeoQuest’s view of Open Spirit is slightly less bullish. While SeisClass has been released as a ‘true’ Open Spirit compliant application, PDM understands that this is not the same as the GeoFrame version, in other words, Open Spirit is not the interoperability mechanism used generally in GeoFrame 4. An insider confided to PDM “Open Spirit is GeoQuest’s preferred mechanism for integrating and interoperating with our competitors’ products.”
© Oil IT Journal - all rights reserved.