Enterprise Services
Showing results for 
Search instead for 
Do you mean 

Information Management for Better Decisions

DaveBennett ‎02-19-2014 04:12 PM - edited ‎09-30-2015 07:01 AM

I’ve spent the past few weeks having some lessons regarding the importance of information management and governance being heavily re-enforced across drilling, geological, and HPC data sources.  It starts with a need for data to make intelligent decisions and branches into some unlikely locations for source information.  It highlights the fact that better data drives better decisions, improves revenue, reduces down time, and avoids incidents.  The key is having the appropriate data available in a timely manner to support effective decision - and having access to both the data and the decision maker at the needed location.  As simple as this concept sounds, business operations and their associated results indicate many organizations struggle to achieving.


As a case in point, in 1979, three major failure modes were documented for NASA’s space shuttle program.  These issues were addressed with appropriate mitigation strategies that were identified, documented, and integrated into program flight and support procedures.  In 1986 the first of these failure modes resulted in the loss of the Space Shuttle Challenger and its crew.  In 2003 the second of these failure modes resulted in the loss of the Space Shuttle Columbia.  These two significant incidents highlight how processes, procedures, and information required to support decision making can degrade, be modified, or get bypassed as organizations change over time. 


In our example, the processes for avoiding both incidents were identified and documented prior to the Space Shuttles inaugural flight in 1981.  Between 1981 and 1986, events and activities transpired that separated critical information regarding launch procedures from decision makers.  Under perceived time constraints and organizational pressure, decisions were made with limited information by people who had not helped develop the original program.  The combination of limited time and lack of access to knowledge caused a shut down in the American Space Shuttle program for over two years while information and procedures were updated.  This process repeated itself in 2003 with the Space Shuttle Columbia, with Space Shuttle flights resuming after another lengthy review. 


These incidents reflect deficiencies in best practices for management and governance of data.  In most situations, we find traditional data management and governance providing nominal methods to ensure data is available to support decision making.  However, these methods and systems consistently break down and fail when revised, when faced with unusual exception criteria, or when not maintained.


Some exception criteria are consistent and known.  For example, at multiple clients I work with, fire and safety procedures are reviewed each time a meeting starts or a new facility is visited to validate all participants are informed on proper procedures should an emergency occur.   The exception criteria for fire procedures are addressed up front as both a precaution and a reminder to maintain a safe environment.  Other exception criteria are less visible.  A frequent issue that crops up is product quality and implementation.  In a steady state operation we expect data, management, and processes to remain within a consistent range of parameters.  When exceptions are introduced into the system, such as part that does not meet specifications or is consistently borderline after a manufacturing change, but still meets the specification, we find work processes are subject to breakdowns.  


If we look at Space Shuttle incidents, it can be observed that that managing by exception criteria or accepting a change without re-examining the complete system fails over time.  Issues that create failures include changes in personnel, in production processes, in documentation procedures, in how information is archived or removed, in how data from acquisitions is integrated, how users adhere to tagging and metadata processes, and how this supports master data schemes.  All of these processes address information identified as critical but typically fail to address how information is included or excluded in the critical category, where it falls in the range for that category, or fails to periodically review what data is considered critical from a business and process point of view.  


As an additional complication, traditional enterprise level data governance processes are being superseded through the Internet experience.  Users and decision makers want data they need available on their smart phone, delivered with sub two second response times after searching across petabytes of unstructured and untagged data.  So what capabilities at what cost becomes the question?  Internal information management works with business to measure value against costs, including considerations for time saved during data acquisition and the value of improved decision making.  This then gets measured by users who compare the accuracy and speed of internal sets of data against free or paid sets of data available from the internet. 


As a case in point, many Oil & Gas clients I work with don’t find nirvana with an internet based parsing solution, while most users don’t comprehend the difficulties in securely managing and delivering value from internal data repositories.  With these repositories spanning structured and unstructured sources – including sensor and measurement systems, most analysis require consideration of targeted aspects or problems, forcing the data to be rotated and analyzed through a differing set of strategies.  Results from these analyses then provide input into monitoring and management systems to support performance optimization, or forward looking predictions.  To complicate data assembly, information may be in a set of text files, pictures, charts, videos, voice, and database files necessitating clean up before it can be used in an analysis process.  This takes us beyond Hadoop parsing schemes for unstructured data sources and requires us to consider how data located in this manner may be integrated with structured data stores to analyze production operations. 


If nirvana is having a mobile device securely provide the decision maker in the field with a combination of operational, maintenance, and production data so that a quick, effective decision can be made, an adaptive information architecture must exist to enable it.  For my Oil & Gas clients, this target data it is most likely information that helps them quickly make decisions regarding drilling, production, or refining.   It may be as simple as a red, yellow, and green light combination that identifies safety issues or optimal production ranges, or it may support drilling down into operational logs, maintenance records, and equipment issues. 


Where this data comes from, and how it has been filtered, regardless of format, is critical.  Accurately maintaining the value chain this represents through organizational changes, introduction of new products, and evolving business priorities can be difficult, but is required for accurate timely decisions.  All of this is compounded by the explosion of available data to process, the perceived ease of use associated with internet based searches, and the helpful but sometimes distracting data that accompanies them.  Managing the challenges to maintaining this value chain while leveraging the awareness provided by the now available and accurate data, requires a commitment to managing continual change.  This may be through a process that continually tests information models, provides business validation, or revisits data models and definitions in structured and unstructured environments. 


Helping manage and govern data while leveraging both structured and unstructured data sources in support of faster and better decision making is one of the solutions HP can provide to help improve results for Oil & Gas companies.  For more information regarding HP’s Oil & Gas industry capabilities, please visit hp.com/go/oilgas. 




0 Kudos
About the Author


Nov 29 - Dec 1
Discover 2016 London
Learn how to thrive in a world of digital transformation at our biggest event of the year, Discover 2016 London, November 29 - December 1.
Read more
Each Month in 2016
Software Expert Days - 2016
Join us online to talk directly with our Software experts during online Expert Days. Find information here about past, current, and upcoming Expert Da...
Read more
View all