Tena Allain heads up the American Petroleum Institute’s (API) ComProServ task group which has been established to investigate standards for financial transactions. ComProServ standards support the specification and execution of complex products and services. Allain told PDM how ComProServ’s twenty member companies have come up with XML-based standards for thirteen transactions, along with a common library and schema.
ComProServ works to describe complex services such as well casing programs or cement jobs at a data and work order level. Such information will then be traded back and forth between operators, contractors and suppliers. ComProServ began by looking at work done in the chemical and electronic industries (RosettaNet) and plans to re-use this work.
The API work has its roots in previous EDI-based standards, used for invoicing of drilling and geophysical services. The EDI standards are free-form text based so the move to XML allows for more structuring of information, more context and metadata to be included. According to Allain, most US majors use the PIDEX EDI standards for services and joint interest billing, but there was less take-up for these standards in the smaller independents. The move to XML may increase take-up for the smaller players.
If it ain’t broke
The ‘migration’ of EDI based standards that are in everyday use to embryonic XML standards is a highly charged subject. As they say in Texas, ‘if it ain’t broke, don’t fix it.’ Allain said that EDI is still going strong. Many users do not even deploy the latest versions of the EDI standards. The advent of low cost bandwidth has significantly reduced the costs of EDI transactions. On the other hand, XML is not all that cheap – particularly when the costs of an Enterprise Application Integration platform such as Tibco is included in the calculation.
BP’s Roger Lott heads up the European Petroleum Survey Group (EPSG). Lott told PDM how the EPSG started out as an informal meeting of senior survey personnel from the major North Sea operators. In the mid 1980’s, the group pooled internally compiled data on worldwide coordinate systems, which grew into a significant resource, which was leveraged by POSC in the Epicentre data model.
EPSG standards have seen take-up outside of the oil and gas industry – particularly by satellite image specialist SPOT which rolled the POSC/EPSG work into the GeoTIFF standard for geodetically referenced bitmap imagery. The EPSG work has been further leveraged through the Open GIS Consortium, extending the EPSG’s work into the Geographical Information System community at large. The EPSG data set is now a de facto worldwide reference and is available online as an Access data base from the EPSG website.
The EPSG currently working on a new version of its specification which will become the ISO 19111 geodetic standard. This will be available in the next release of the EPSG database – Version 6.1. The EPSG plays something of a proselytizing role with respect to the GIS industry – in the hope that promotion of the EPSG standards will help its members. Other EPSG work includes the issuance of ‘guidance notes’ – best practices for the positioning business.
Jack Gordon (Conoco), who chairs the Geoshare user group told PDM that “Geoshare is still in use, judging by the 3000 hits per week to the Geoshare website”. Most visitors are looking at the data model itself. Geoshare is a stable data transfer environment with few requests for changes. There have been proposals for an XML-based version of Geoshare, this is still under discussion. Gordon believes that the current Geoshare model, based on the API RP66 protocol, still has merit. The standard offers a sophisticated environment, with built-in support for units of measure (UOM). It is efficient for binary data and is ‘self-documenting.’ But Gordon acknowledges “XML is a wonderful tool – it is great to be able to read the ASCII information.” In any event, future XML standards should build on prior work such as Geoshare’s UOM standards.
Finder to OpenWorks
One significant usage of Geoshare is to connect Finder to Landmark’s Open Works. An Excel mapping of Finder and OpenWorks attributes is available on the Geoshare website. Who uses Geoshare today? According to Gordon, Conoco does along with Anadarko and Burlington. From Schlumberger’s ongoing support for the standard, one suspects that the technology still has currency within GeoQuest, but if there are any big Geoshare projects out there, everyone must be sworn to secrecy! Geoshare is calling for papers for the April PNEC conference (see page 10) which follows the AGM.
LAS chairman Kenneth Heslop told PDM that the version 3 of the Log ASCII Standard (LAS) is currently under consideration by several administrations as the required standard for data submission. LAS 3.0 handles well logs, core analysis, well tests, and deviation surveys. One current LAS project involves creating a new code system to replace the old API codes and content rules. The new log codes incorporate up to date tool mnemonics and will provide for new codes as new tools are developed.
LAS 3.0 content rules let government regulators and users specify minimum acceptable content. The LAS ‘Certify’ program will be enhanced to check an LAS 3.0 file to ensure compliance with both the LAS standard and specific content rules. Work on these two aspects of LAS 3.0 will continue in 2002. A Windows version of Certify is now available for download from the website. This version supports all versions of LAS, and has expanded checking and reporting. Future versions of this application will also support content rule compliance.
Open Spirit CTO Clay Harter gave PDM an update on Open Spirit’s activity. But first we popped the question: Is Open Spirit (OS) really a standard? Harter points out that OS publishes business object definitions on its website. While these “could be implemented” by a third party, Harter admitted that without the OS license and runtimes from Landmark and Schlumberger, this is unlikely to be a practical proposition. The drift away from the POSC business objects standards began when OS developers realized that the division of labors between specification and deployment was not tenable. Harter summed up “‘Standard’ is a bit misleading – OS is rather an ‘interoperability solution.’ We would love it if OS became the de-facto standard for upstream interoperability.” OS is talking to POSC about finding an appropriate way of presenting and describing these aspirations.
OS has ‘borrowed heavily’ from POSC Epicentre and Business Objects and still believes that developing and evolving business object standards should be an open process and that other companies and standards bodies should be involved. This might involve cycling new business objects submitted by OS developers through a POSC-like approval process.
OS Release 2.2 introduces new domain objects and extends current objects to the latest releases of GeoFrame and Open Works. This buffers OS users from changes in the underlying data stores thanks to a ‘late binding’ interface. Late binding uses the same techniques as the dynamic link library (DLL), prevalent in the Windows environment.
OS 2D grids will be extended to tri-mesh and point sets. Object richness will be extended in the subsurface space and into drilling and production arenas. OS is looking at Microsoft’s .NET technology to leverage legacy (UNIX) data stores. OS is also working with ESRI to interface with ArcView.
POSC chairman David Archer told PDM that 2002 should see a renewal of interest in POSC’s flagship database Epicentre. This centers around work being done for Statoil (see page 3 of this issue) and for the China National Oil Co., big Epicentre users.
POSC’s XML work is ‘maturing’ – with a move to reusable XML ‘components.’ POSC has worked with PPDM on the joint reference values project, resulting in an agreement with PPDM for a cross-posting of related URLs on organizations’ websites. POSC is also working to ‘reduce confusion’ with other workgroups and is sharing personnel with Geoshare. Phase I of the Practical Well Log Standards was completed in 2001. Phase II will continue into 2002 and involves the rationalization of mnemonic data. The Shared Earth Model is now an EU-funded research project, with work being done by the IFP and PDS.
Reflecting on POSC’s evolving technology focus, Archer said, “We have been through generations of technology – component, virtual data stores, unbundled and distributed architectures. In fact CORBA is still too tight – things need to be unbundled further – especially the stuff that we do not control such as security and authentication. All this calls for a web services approach, leveraging technologies like XML, SOAP and WSFL. Another tendency is the move towards Open Source – we are making the POSC Universal Units Converter available as Open Source.”
The planning process is ongoing, but the current focus is on architecture – the framework – as well as content. Well Header ML is likely to be leveraged by CDA and the MMS. Development continues of Well Log ML (which ultimately should replace the LAS) and Production ML will be extended to incorporate joint venture reporting – this has application both within the enterprise (Shell is looking to use this internally) and for regulatory reporting. Other plans include looking into how a well lifecycle approach could keep track of well data throughout its history. The plan is to use something analogous to an automobile chassis number to uniquely identify a well, and to track associated data as it moves between partners and contractors. More work will be performed internally using web references services to provide a data model dictionary – with dynamic updates of POSC data model changes available on the web. POSC may develop its own Petroleum Industry Data Dictionary (PIDD) with an evolving record of well log mnemonics, and industry reference data. POSC has had a difficult year, especially with a reduced member base due to mergers. These have brought some newcomers into the fold – such as Paradigm. The Indian ONGC also joined in 2001. The relationship with PPDM is aimed at “avoiding duplication and conflict – by ensuring that both organizations are aware of what the other is doing.”
PPDM President Scott Beaugrand told PDM of his success in bringing Saudi Aramco into the PPDM fold. Aramco’s data modeling and application development group is looking to standardize the data model for use with off-the-shelf software. The plan is to ensure that the 50 years of Saudi data will be accessible in a standard, non-proprietary data model.
At the PPDM AGM this year, Woodside Petroleum presented the results of the first ‘real-world’ implementation of the PPDM Spatial I project. Spatial I uses an intersection table to decouple metadata in PPDM form from a generic spatial database. A new PPDM-related project, ‘Spatial II’, has been initiated by ESRI. Spatial II will build a PPDM-based ESRI ‘geodatabase’ for the oil and gas industry. Like POSC, PPDM is working to leverage its data modeling work with XML, and has registered XML data exchange specifications with the xml.org repository. Beaugrand believes that there is increasing recognition of the importance of data modeling and that funds available for good projects. Data exchange, software applications and other services based on the PPDM model have reduced user and supplier costs and improved the effectiveness of information technology.
Activity in PPDM’s core business area of data modeling continues with the release of PPDM V3.6 this month. The new release adds to the seismic and information management modules and enhances the stratigraphy module. New additions to the seismic module provide for lifecycle data management, from acquisition through processing and interpretation, including transactions and brokerage. Data encryption, error tracking and tape copying are now supported in PPDM Version 3.6. Entitlements to data, information and products are described in a new module and cost center summaries and pointers allow users to integrate their business and financial databases effectively.
The Society of Exploration Geophysicists (SEG) Technical Standards Committee (TSC) pretty well lays down the law in terms of how geophysical data gets recorded. SEG TSC chairman Alan Faichney of Concept Systems provided this update. The new SEG-Y format (Rev. 1) incorporates multi-component data and integrates the European Petroleum Survey Group (EPSG) geodetic reference data. A final draft specification was published in the Leading Edge in August 2001. Subsequent to this publication, one astute reader pointed out that the multi-component specification omitted multi-component source data. This issue is to be addressed by the committee, and a final revision will be out shortly. Faichney notes that the ‘tone and structure’ of the new revision has been accepted.
The Shell Positioning Format (SPS) is aimed at land seismic acquisition and solves the problems inherent in using P1 with multi-component acquisition. SPS is widely used by recording system manufacturers and software companies involved in land acquisition. SEG-D is also under review, although again this mainly concerns equipment manufacturers as all data is now transferred to SEG-Y immediately. The new SED-D will facilitate recording data direct to disk. GXF is a new gravity and magnetic format for gridded data. It has emerged as a de-facto standard and will likely be accepted as a recommendation. Again, the EPSG positional resource will be embedded. Faicheny, as SEG TCS chair, has been co-opted onto the EPSG.
The original Well Information Transfer Standard (WITS) specification evolved out of Amoco’s Critical Drilling Facility, Chevron and Mobil’s Data Centers, Statoil’s Drilling Automation and Real Time and other proprietary initiatives. The original WITS format is binary, efficient but not very portable. John Shields (Baker Hughes Inteq) presented a new initiative, WITSML at the POSC fall member meeting in London. BP and Statoil got together to update the WITS standard for ‘right-time, seamless flow of information’ between operators and service companies. BHI, Halliburton and Schlumberger are participating. WITSML will connect to OpenWorks and GeoFrame through an application programming interface (API). In January 2002 there will be commercial offerings from the major service companies and at least one smaller software house. WITSML is web-based, and follows the W3C XML guidelines. This was a ‘good decision’ in terms of web services - it is platform and language independent. The API can be prototyped with Visual Basic. Shields noted that the logs and real time objects are not very standard XML. WITSML ‘looked at’ POSC WellLogML and tried to be ‘close to’ POSC. The real time specification is ‘self-configuring and self-describing’.
Standards on the web
Log ASCII Standard (LAS)
American Petroleum Institute (API-PIDEX)
Society of Exploration Geophysicists (SEG)
Well Information Transfer Standard
Open Spirit Corporation
European Petroleum Survey Group
Petroleum Open Software Corporation
Public Petroleum Data Model Association
The Oracle ‘Synergy’ project (PDM Vol. 4 N° 5), once a major Epicentre development, is now deceased, reflecting both Oracle and Statoil’s changing priorities. BizTech (formerly COM) for Energy appears moribund, at least judging from its public website. Interestingly, both these projects were heralded by their protagonists as a new breed of service-company funded standards initiatives with more focus and fiscal stability! On the product data/construction and engineering front, the POSC/CAESAR organization also appears to have gone extremely quiet. This reflects both the difficulty of replacing proprietary technologies by standards and the hegemony of the large construction companies.
This review was abstracted from a Technology Watch report by The Data Room. The Data Room produces around 15 in-depth Technology Watch reports per year, along with a detailed annual summary. For more information on The Data Room’s Technology Watch service email email@example.com fax +331 4623 0652 or call +33 1 4623 9596 .
© Oil IT Journal - all rights reserved.