Service company IT and corporate IT
Friday, June 29, 2012
Who is responsible for fitting together IT from service companies and the corporate IT department? The business unit managers. Are they the best people to do it? By Mark Reynolds, staff drilling analyst, Southwestern Energy (Houston)
Why do companies buy computers? Such is the question presented to my freshman computer science classes.
After much prodding and discussion, the students get it - to make money.
Paraphrased, why do oil and gas E&P managers invest in technology? Again, to make money.
Sometimes making money may result from reduction of injuries, reduction of spills, or reduction of NPT.
Other times, monetary value is realized through boreholes which are straighter, drill faster, and stay in-zone. In other cases, technology improves communication and asset management.
One thing is clear: managers invest in technology which is operations-centric and bottom-line focused.
Operations-centric systems and technologies (specifically software and data systems) are not financial systems and are not accounting systems; they provide explicit functional benefit to the operations side of the enterprise.
Drilling systems and drilling technologies are distinctly focused on keeping the bit turning - reduced HSE incidents, reduced NPT, increased wellbore quality and increased ROP.
Business unit challenges
Development, implementation, and management of operations-centric [software and data] technologies is not well understood either within or without of the business unit.
And the business units must locate solutions: service companies, corporate IT, or business unit (in-house, in-group) solutions.
Service companies provide many solutions, often a systems-wide solution.
Corporate IT departments develop the requisite application to custom specifications, often integrating in-house data and applications.
Business units solve problems - quickly, internally, and with differing degrees of success - system solutions and integrated data solutions.
Industry standards have helped smooth the dissimilar edges of diverse subsystems. WITS, WITSML, and PPDM (most especially WITSML) contribute to a level of interoperability.
Furthermore, these standards form the glue between the subsystems far superior to emailed spreadsheets and 24-hour reports.
But, in many cases, the advantage has been lost due to the non-standard nature of the standards as well as the increased complexity of the drilling problem.
Service companies offer many timely and out-of-the-box solutions. And these solutions are typically designed with a level of forethought, planning, and agility but they must meet expectations from a range of potential clients. Thus they are seldom a single solution with a great fit.
Systems which provide a high degree of customization will achieve a degree of agility, but with the cost of a level of complexity.
Nevertheless, solutions provided by service companies are critical to the drilling challenge. Drilling engineers have grown adept at shoehorning diverse, yet synergistic, service companies into a functioning unit.
Rig / PVT sensors and subsystems, downhole equipment and subsystems, mud logging subsystems, and geologic subsystems can be brought together to produce a system of values - a system of visualizations - which can be assimilated by the engineer.
And this has been successfully achieved by all successful E&P managers.
Traditional IT departments
Traditional IT departments want to take over, manage, audit, and control systems.
They approach the problem from the viewpoint of a technologist - a technologist bound by a decade of regulations and audits: Sarbanes Oxley, ISO-9000, and various SEC regulations.
IT departments understand analysis, development, scalability, and roll-out but product development cycles are seldom nimble and responsive. Agile and Scrum development techniques have improved the situation, but not sufficiently improved to keep the bit turning in today's systems.
Ad hoc fitting these solutions together - solutions from the service company and from the IT department - has become the challenge of the business unit expert.
Immersion within the drilling environment promotes an agile and responsive mindset, but discourages robust designs.
Thus the business unit operations and processes tend to spring from spreadsheet solutions designed to address 1) immediate challenges and 2) interconnecting dissimilar systems. Thus, business unit solutions are frequently non-integrated technologies.
But spreadsheets have accomplished unbelievable feats; but they are inherently ad hoc solutions and normally require hands-on nurturing to keep running.
Move to closed loop
As the industry moves away from hand-on-the-brake drilling mentalities into closed-loop automated drilling, look ahead predictive analysis, and nearby well comparisons, the ability to consume and process ever increasing datasets becomes the driving factor.
Support-center based performance visualization is becoming more common place thus raising the bar in regards to monitoring. Specifically, traditional threshold based alarms are giving way to knowledge systems, case-based reasoning, and closed loop systems.
Business unit based teams of data analysts, data engineers, and systems architects become central to the new drilling paradigm.
This Business Unit IT may exist as a business unit organization with connections to IT or may exist as an IT extension tightly coupled to the business unit. Either way, rapid-reaction operations-centric solutions are required.
Regardless, Business Unit Development teams create intrinsic value in the drilling datasets by moving away from 24-hour morning reports and into the real-time data streaming environment.
And drilling challenges may force the real-time streaming systems to move from 10-second update rates down to 5-second, 2-second, 1-second, or even sub-second updates.
System and subsystem interplay cannot be understood when the sampling granularity is too large. The ability to watch and analyze trip speed, vibration, and most transient events is non-existent with updates slower than 10-second rates; 2-second or even 1-second updates are preferable in many scenarios.
Service companies and solution providers are rapidly developing new integrated solutions. Case-based reasoning, complex event processing, and closed-loop control provide new levels of centralized direction, steering, and control. But bringing these new services (internally or externally developed) requires a new level of industrial and software management expertise.
Management oversight is asking for more complex analysis of more data. Big data systems, business intelligence, and traditional data mining become the ingrained minimum expectation.
When coupled with increased asset scheduling and mobilization expectations, with the increased governmental permitting and reporting expectations, and with increased management oversight, the integration of current drilling parameters with asset management becomes its own challenge - and this challenge can only be understood and solved by the ingrained / integrated business unit team of data analysts, data engineers, and systems architects.
But technology must increase the ROI; it must increase the generated revenue or decrease the capital expenditure. And the business unit team must, themselves, produce a net increase in company value.
Quantification of the ROI benefit is difficult at best and unwieldy in most scenarios.
For example, asset management and scheduling ROI can be inferred by idle rig-days and reduced governmental penalties. Operations-centric borehole control ROI can be assumed and quantified based on 1) reduced sidetrack counts in known areas, 2) borehole straightness, in zone-ness, and achieved lateral length, and 3) higher production volumes.
Quantifiable justification of the business unit embedded team of data analysts, data engineers, and system architects when compared to internal IT organizations as well as service companies is no less challenging.
Comparison of raw estimates to perform a project (time and head count) is meaningless unless a demonstrable accurate task estimation system is in place - but this frequently precludes agile and nimble development teams.
The business unit development team, when properly formulated, must address the drilling challenge in an embedded and integrate fashion and must consist of data analysts, data engineers, system architects, and a management layer adept at creating the operations-centric, agile and nimble, service group.
The business unit development team can, when properly constituted, equipped, and encouraged, become the E&P manager's go-to asset.
So the case for business unit development teams becomes a management approximation of the team strengths, the business expectations, and the cross-discipline need for results.
The expectation for success of the business unit team must be compared to internal IT teams and to service company turn-key supplied solutions.
When expectation for business unit teams to respond to ever tight tolerances to time, production, and HSE, the case for the business unit, operations-focused, development team becomes more readily apparent.
Business unit software and data systems development teams can, and will, make money for the organization through improved and faster drilling, improved production, and reduced HSE, NPT, and asset downtime.
Mark Reynolds is currently at Southwestern Energy (Houston), where he works in the Fayetteville Shale Drilling group as a Staff Drilling Data Analyst. In this position, he pulls his experiences in data processing, data analysis, and data presentation to improve Southwestern Energy's work in the natural gas production and mid-stream market.
Mark began developing military avionics systems for General Dynamics and Sikorsky Aircraft. Since 1990, he has been developing Systems and Applications for the Energy Industry including integrated information systems, systems analytics, real-time processing, and operations management.
Any opinions expressed in this article are personal opinions of Mark Reynolds, and are not related to any business activities of Southwestern Energy