Stephano Ventura described ENI’s reserves information system, a common data environment for ENI’s financial reporting. A central database in ENI’s head office serves 20 geographical business units. ENI was able to preserve existing systems by ‘wrapping’ them into an E&P data warehouse. Different data reporting formats are transformed on the fly. Building blocks include an Oracle repository, an Apache Tomcat web server and Business Objects’ reporting tools. End users connect across the WAN with a web browser or Microsoft Office applications—leveraging Excel OLAP Services.
A user group has been created to address cultural issues and customization to local reporting requirements. Usage monitoring allowed for fine tuning of the system’s functionality. The project was described as successful, but it was a ‘long and painful process!’ The project involved ‘limited’ outsourcing. Ventura is a strong advocate of in-house development on a project of this complexity.
Paul Cleverley (Flare Solutions) provided an update on information delivery to Shell’s new business, exploration and development teams. The idea is to use all available information in decision making and not to waste time assembling it. Planning for information re-use mitigates information loss when a project is put on hold. The solution, which is global to Shell, was designed to be easy to use. For Cleverley, ‘If you need to train, it is not an end-user tool.’ Cleverley stresses that straightforward full text search, as used by many oil and gas companies limits retrieval accuracy. Shell’s best practice is to constrain text search with taxonomies to ‘pin down’ terminology for assets, production and E&P data types and project names. The system also leverages Shell’s metadata stores and search ‘helper’ applications like MetaCarta. Enabling technologies include Schlumberger’s DecisionPoint and of course, Flare’s Catalog (OITJ Feb 07). ‘New knowledge’ can be derived from usage patterns.
The system offers fine grained control over results’ entitlements that makes sure confidential information stays that way. External data sources such as IHS, SPE, CASP, Telus, K-Res and C&C can be blended into the search results and visualized using Shell’s ‘Bright Sky’ knowledge map application. The system has an estimated 4,000 users. The system is replicated across Shell’s worldwide IT hubs with ‘cloned’ metadata. Deployment was performed by IBM using WebSphere-based automation. In the Q&A Cleverley revealed that the system was originally delivered as an SAP I-View portlet. But there was little take up for the Portal and today the system is a stand-alone application.
Maintenance performance analysis at BP’s production facilities was the subject of Ian Hendry’s presentation. BP has developed a tool for operations maintenance of onshore and offshore facilities that now has 4,500 users and one million ‘maintainable’ items. BP invests some £3.6 million per annum on data collection,. Maintenance operations are highly dependent on data quality. Hendry was a strong advocate of IBM/MRO Software’s Maximo product that underpins BP’s system. BP has the same issues of data quality maintenance as Shell.
Lars Olav Grovik (Hydro) explained that in Norwegian, the acronym DATA can be interpreted as ‘double work for all.’ A revamp of Hydro’s drilling and G&G software portfolio with the deployment of end user-centric tools like Schlumberger’s Petrel has contributed to a reduced focus on data management. Hydro’s data management was strong in the past, but is challenged, not by a lack of technology, but by a lack of manpower. A data management program kicked off in 2006 with a focus on end-user efficiency. ‘Data management should accept nearly anything’ and provide fast and simple access to data that is close to the end user. Hydro’s involvement in integrated operations has impacted the program with a requirement to enable cross silo data flows. Grovik observed that Exxon, BP, Shell and Statoil have very different policies. There is no ‘right’ way to do it. The issue of data mining has proved problematical. Hydro is keen to offer a simple GIS interface to all of its data and is likely to leverage Google Earth. If you can’t get the results of your study into Microsoft PowerPoint ‘it’s a big problem.’ This has been a particular issue for geological models.
© Oil IT Journal - all rights reserved.