Judging from presentations at the IRM-UK Master Data Management conference held last month in London, the MDM problem is easier to state than to resolve. Andrew Schulze outlined proof of concept MDM initiatives in Shell. Both SAP and Microsoft’s MDM tools have been trialled, with help from Accenture and Wipro. The tools were assessed as master data ‘peers,’ and evaluated for integration with Shell’s information architecture (Oil ITJ January 2011). Scripts for various scenarios were run on hardware provisioned through Amazon web services. A gap/functionality scorecard has been established and today, seven data objects are live, six projects active and three PoCs complete. Key finding is that MDM needs to be delivered for the enterprise. Responsibilities must be clearly defined ‘so governance does not fall through the cracks.’
Chris Bradley (IPL) and Eldar Bjorge outlined Statoil’s model-driven approach to MDM. Internal metrics showed that 85% of incidents at Statoil’s supply chain service center are due to master data errors. Problems arise from poor master data within and across environments and from silo-oriented, stand-alone applications. MDM is being driven by Statoil’s need to rationalize drilling, well, lease and supply chain data and to support marketing, processing and renewable energy. A pilot covering enterprise content management of location data has allowed for testing of MDM governance workflows and tools.
One issue with MDM is that projects can’t wait on master data delivery. The answer is to supply ‘just in time’ subsets of master data to projects as needed. ‘MDM is not the field of dreams.’ In E&P, managing complex data types involves compromise and the art of the possible. Statoil has established decision criteria for its approach and architecture along with a conceptual model for subject areas that will be put under MDM control. The geography subject area is finalized. Priority and sequencing for other subject areas has yet to be decided.
Jan Henderyckx (Envizion) and Greet Krekels described GDF Suez’ vision of information as a corporate asset—protected through its life cycle, providing trusted information to all stakeholders. GDF Suez uses the InfoSphere toolset to map its information landscape across real-world assets, like rotating equipment, into business and implementation semantics. An InfoSphere business glossary browser contains master definitions. Definition ownership is thrashed out by a coordination board and subject matter expert task forces working across divisions. Multiple models are built and maintained in InfoSphere data architect. Henderyckx observed that low maturity in data modeling and poor standards enforcement results in a strong business case and lots of ‘low hanging fruit.’ More from IRM-UK.
© Oil IT Journal - all rights reserved.