BP is looking to implement a standard data architecture for real-time drilling & completions information. This will enable us to make wider use of our collaboration centres through a common approach and will allow us to leverage the developing WITSML standard more effectively in our operations. BP uses many different service organisations to deliver its global drilling & completions agenda. In the past, this has resulted in a lack of standardisation in real-time information flow. We have been unable to share data and expertise readily between different operating centres. Applying a common approach to information access on a global basis will enable us to streamline our operations and make wider use of emerging collaboration technologies. WITSML is now finding greater uptake in our industry and we are seeing new compliant intelligent software tools emerging from different vendors. These will undoubtedly enable us to drill and complete smarter but we need a standard information infrastructure to deploy them widely and consistently. It is these smart tools that will ultimately deliver real value to oil companies. The adoption of a standard D&C information architecture may seem an obvious step but there are very few oil companies that can claim to have delivered it. WITSML is the vehicle to allow this to happen and over the next few years it is likely that there will be much more emphasis on open connectivity between different vendors and services supporting the drilling & completions functions. 1. Introduction The rapid development of digital technology is touching all disciplines in the oil industry and is providing exciting new opportunities and capabilities in areas such as automation, remote monitoring, simulation and enhanced visualisation. In terms of capital employed in digital technology, there can be few functional areas that have the potential to deliver a higher return on investment than drilling and completions due to the sheer magnitude of today's operational costs. A saving of at least a few percent in non-productive time should be readily achievable for most oilfield operators and service companies with more effective use of real-time data to improve operational efficiency and the productivity of scarce, highly qualified discipline specialists. Data capture from drilling rigs has for a long time been the responsibility of the service contractors (typically the MWD/LWD contractors). Their task has been to monitor information real-time on the rig and increasingly at remote onshore data centres. The oil company has generally had the option to have a real-time WITSML feed from the service contractors' data centres and has been able to monitor offshore operations using a proprietary real-time viewer. Typically, at the end of the drilling and completions operation, the oil company is supplied with an archive of the real-time information on a portable media such as tape or CD-ROM. How effectively this information is then used varies widely across the industry. When this approach was first introduced, it was excellent as it represented a significant step forwards from paper logs and other manually gathered data. However, with the expansion of the digital world, the demand for information has grown significantly and a number of opportunities now exist to make better use of real-time information:-Remote operation and support at oil company collaboration centres.Smart algorithms to improve operational efficiency.A greater desire to capture lessons learned and to influence future operations.Integration of all data sources from the rig including MWD/LWD, weather, anchor, positional, etc.

Historically, the drilling industry has utilized real time data (RTD) for improving drilling efficiencies, documenting occurred events, and reactive decision-making. RTD should promote situational awareness with respect to wellbore instability and hazards. Drilling trends identifiable in RTD change over time as abnormal trends leading to hazards and rarely occur instantaneously. These destabilizing events such as cuttings loading, barrier compromise, kicks, pack-offs, wellbore collapse, or other mechanical and pressure-related causes, whether in rotating or flat-time operations, can be avoided or mitigated. Accepting these hazards as a necessary outcome of drilling operations can be a dangerous proposition and denies risk accountability.Enhancing the systemic approach to process safety represents a paradigm shift in the industry and requires changing behaviors through proactively managing and controlling the pace of the drilling process. A new paradigm suggests the use of a deterministic, physics-based model of real time force-balance in the wellbore during the construction process. Identifying the trends of these forces become predictors of hazards. Mitigating and correcting the trends, in real time avoids process safety failures and consistently results in improvements in non-productive time (NPT) and removable lost time (RLT), notwithstanding the large associated costs. The drilling trend behaviors that predict hazards can facilitate leading as opposed to lagging Key Performance Indicators (KPIs). The new paradigm suggested enables hazards avoidance and thus significantly reduces the costs of drilling. This paper describes and explains these predictive techniques as well as offers suggestions for industry training with respect to surveillance and hazards avoidance using RTD simulators, much as airline pilots would use for continual training. 2 OTC-26935-MS • 12 days of Wasted Time, • $ 20,000,000 plus cost of sidetrack. 8 OTC-26935-MS

Management Real-time data is not about well control, it is about well control avoidance. Recent catastrophic blowouts have underscored the value of real-time data and, more importantly, they have also underscored the value of having the right kind of experience to understand well data interpretation in real time. What is the well telling us? How do we use real-time data to ensure a stable wellbore? Real-time monitoring integrated with rigorous total well control analysis is required to embrace and achieve continuous improvement and ensure the safest possible environment. Next generation monitoring requires a step change that includes hazards avoidance as a precursor to drilling optimization. Real-time data can be used effectively to avoid, minimize, and better manage drilling and completion operations. They can also provide the foundational support to improve training in the industry as well as develop hands-on simulators for hazards avoidance. The fundamental definition of process safety is that of ensuring containment. In the case of drilling and completion operations, that means well control. Process safety management (PSM), a regulation promulgated by the US Occupational Safety and Health Administration, is an analytical tool focused on preventing release of any substance defined as “highly hazardous.” This concept can be extended to oil and gas well operations within the context of ensuring that the well is constantly under control. The industry has pursued personal safety, which resulted in significant improvement in preventing accidents commonly referred to as “slips, trips, and falls.” Though this is a notable accomplishment, it had no influence on well control episodes, and blowouts remained a frequent occurrence. As well complexity increases, well control events increase disproportionally. Even the term well control, which should be related to the performance of oil well operations in a controlled manner, became more associated with the operations necessary to fix a situation once control was already lost. Often, the industry depends on blowout preventers as execution tools rather than the fail-safe tools they were designed as. If a driver constantly slams on the brakes of a car at high speed, eventually the brakes fail, or at the least, fail to operate in a timely manner or accurately. Real-time data offers the ability to ensure process safety. Fig. 1 shows the relationship between real-time data and its impact on operational outcomes.

When analyzing root causes for minor or major problems occurring in oilwell drilling operations, investigators almost always can track past events, step by step, using recorded data that was produced when the operation occurred. In recent catastrophic blow-outs, investigators were able not only to determine the causes of the accidents but also to indicate mitigating actions, which could have prevented the accident if they were taken when the operation actually took place. This is a strong indicator that, even though the industry has valuable real-time information available, it is not using it as a tool to avoid harmful events and improve performance. Real-time data is not about well control, it is about well control avoidance. Recent catastrophic events have underscored the value of having the right kind of experience to understand and interpret well data in real time, taking the necessary corrective actions before it escalates to more serious problems. What is the well telling us? How do we use real time data to ensure a stable wellbore? Real-time monitoring, integrated with rigorous total well control analysis, is required to embrace and achieve continuous improvements — and ensure the safest possible environment. Next generation monitoring requires a step change that includes hazards avoidance as a precursor to drilling optimization. Real-time data can be used effectively in operations to avoid, minimize, and better manage operational events associated with drilling and completion. Real-time data can also provide the foundational support to improve training in the industry as well as develop hands-on simulators for hazards avoidance.

No abstract