2011 ECIM Data Management, Haugesund

Upstream data event heard from IBM info-guru Sunil Soares on data governance, Spring Energy on data management in small oils and Statoil’s quest for an information architecture and ‘big data.’

ECIM retained its claim as the premier E&P data management forum with a full house of over 300 attendees—some 35% from outside Norway. In an entertaining keynote, Eirik Newth, who describes himself as a ‘futurologist,’ focused on mobile computing and social media. Mobile computing today is at the same stage as the PC market was 20 years ago. While games like Angry Birds and Word Feud are ‘huge productivity hogs,’ there is as yet nothing revolutionary in mobile, ‘but that will come.’ One problem is lock-in that vendors’ ‘honey traps’ bring—such as Apple’s App Store. Industry needs less ‘killer apps’ and more standards like HTML 5. This should let users use content across multiple devices. Facebook has ‘defined social media’ and demonstrates strong synergy with mobile computing. Facebook’s ‘products’ are its customer, i.e. you! FB is a very futuristic company. This can be good and bad. Facebook may have played a role in the Arab spring, but it can be used against people by the secret services and police forces. Newth left Facebook six months ago. You should be ‘very wary!’

Lars Gaseby (Shell) kicked off the proceedings arguing that it is better to get your data into shape early in the life of a field. If you wait, the well count and data volumes will rise and you will be competing for resources with other activities when production is declining and costs are squeezed. What should you be doing about your data? You need to manage it and assure quality by fixing defects. How? These are the questions that ECIM seeks to answer.

IBM information management guru Sunil Soares observed that ‘governance’ has achieved buzzword status. What is ‘governance? ’ When Soares was writing the blurb for his latest book on the topic (www.oilit.com/links/1109_38), his publisher suggested he ‘write something catchy.’ In fact ‘no one agrees what it is,’ but it is, ‘really about quality.’ For a large organization, the question, ‘how many employees do you have?’ may be hard to answer. You may have to look at different systems (SAP, badging, HR). But then there the question of what is an employee? Staff, contingent, supplemental? Governance involves the formulation of policies for information and data management (considered a ‘technical issue’). Governance in E&P may mean aligning multiple definitions of a well to ‘optimize the information asset.’ Governance spans data quality, standard names for equipment and assets, accounting, HSE, regulatory/compliance. Metadata is the key to tying information across sources via consistent definitions.’

Soares’ discourse morphed into enterprise IT, data stewards and a envisaged a data governance manager ‘sitting in IT.’ The RACI matrix (www.oilit.com/links/1109_39) also got a mention. Comment—Soares’ talk was to an extent, subliminal advocacy for IBM’s ‘Infosphere’ master data platform. Now who might be best set to implement such?

Johan Kinck of Spring Energy described the E&P data landscape from the viewpoint of a small oil company. Smaller companies’ IT/DM function requires a ‘jack of all trades.’ Typically, key geotechnical IT is managed in house, with help from hosted services such as Petrobank, L2S, NPD etc. ‘Commodity’ IT is outsourced. The application environment will likely include tools like Petrel, SMT, FFA, possibly linked together with OpenSpirit. Data quality is just as important an issue for small companies as it is for the majors. Imperfect E&P data standardization is problematic just as is tool diversity. A complex IT landscape spans Linux, SQL, Oracle, Access, and various flavors of Windows. DM capability from vendors is ‘uncertain.’ Vendors are not ‘evil’ although they do have their own agenda. ‘Sometimes we feel a bit neglected.’ Small companies can rarely afford high-end solutions and avoid large investment in hardware or tools. Spring, along with other, smaller Norwegian operators, has set up ‘Information Management Oslo,’ to share insight and experience.

Statoil’s Eldar Bjorge has seen a lot of data management initiatives—from Geoshare, DAEX, POSC to PPDM. These tend to focus on data modeling rather than information management. Statoil’s SCORE project was about an enterprise architecture, before the term was invented. If the oil industry was a leader a decade ago, other verticals have caught up and brought new concepts and powerful resources to play. EIM is the subject of many publications and reports from consulting houses. Bjorge noted contributions from Accenture, Forrester Research, DAMA, Schlumberger and The Open Group. While E&P is different, ‘we have something to learn and something to share.’ Statoil has used the DAMA data dictionary as the basis for its own data management Wiki. Statoil’s IM functions and domains have been mapped into a DAMA-esque framework. Statoil now has an IM governance framework in place and an enterprise data model based on the DAMA framework. E&P has had a long focus on specific challenges, but now other industries have caught up and have stuff to teach us.’

A follow-up presentation from Liv Stordahl Borud fleshed-out Bjorge’s presentation revealing that Statoil was currently rolling out Software AG’s Aris tool (www.oilit.com/links/1109_32) to underpin its information architecture. A master data pilot worked in parallel with the EA project providing input on the information layer in the architecture. Initial IM maps and catalogs are now being migrated from Excel to Aris (www.oilit.com/links/1109_32). The information model includes subject areas and related processes, stewards, data source and master storage.

Lars Olav Grøvik offered an entertaining take on the current state of data management which remains a pain point for many operators. One joint venture reported real oil production from a non existent well. There may be different number of wells in different systems. Looking for data still delays projects and it is the same for most large oils companies. Re-determination and unitization should be easy with modern information systems but they are not. There are always errors in the unitization database. Departments initiate ‘shadow’ data systems because of the challenges of corporate IT leading to reporting and compliance issues. A recent communication from NPD, the Norwegian regulator said that no Norwegian operators were in total compliance! Statoil was in the mid to upper range. The good news is that things are still working despite the data explosion. The not so good news is that data is still ‘exploding’ with even more data generated by permanent seismic arrays and petabytes of seismic on disk. New uses for data mean that we need to break the link between databases and applications. Grøvik sees possible salvation from the ‘big data’ solutions deployed at Yahoo, eBay and others.

Click here to comment on this article

Click here to view this article in context on a desktop

© Oil IT Journal - all rights reserved.