Rio Oil & Gas 17-20/September 2012 Conference – Rio de Janeiro, Brazil, “How we improved operations in drilling pre-salt wells”, Augusto Borella Hougaz (Petrobras), Luiz Felipe Martins (Genesis do Brasil), Carlos Damski (Genesis Petroleum Technologies), Jéssica Lima Bittencourt (Genesis do Brasil), Luciano Machado Braz (Genesis do Brasil).
The development of pre-salt wells offshore Brazil has been one of the most challenging projects in history of E&P. Facing water depths of 2000+ meters, a salt layer 2000 meters thick to reach reservoirs at 7000 meters below sea level, has to use technological and procedural techniques never used before. In those 2 years of exploration of those fields many technologies were tested and improved. This paper describes the systematic approach was taken to analyze, plan and follow-up the development of drilling operations campaign in those fields, and the case study of overall process improvement. The assumption was to apply a risk analysis tool which uses previous data to analyze the performance and plan future time. The goals of this process are continuous improvement of execution and process control for each operation. Looking into previous performance, new interventions were planned more accurately and further improvements were studied. The frequent follow up of the drilling intervention was done using the statistical base to compare the most recent results. At operational level it was possible to see if the last operation was done in the 1st, 2nd, 3rd or 4th quartile of its related statistical distribution, as well as to verify the difference within P10 and P90, which indicates the control of each operation execution process. The same process was done for some rig related operations and for the whole intervention at end of it. Close contact with the intervention progress was kept and actions taken in any major deviation from the plan.
This paper describes the case study where the process control and optimization of the total time for drilling 10 wells with similar design was measured between March 2009 and May 2011. It resulted in significant improvement in the drilling process.
José Ricardo P. Mendes (Unicamp)
Kazuo Miura e João Nuno V. Calvão Moreira (Petrobras)
Carlos Damski (Genesis Petroleum Technologies)
Luiz Felipe Martins, Naisa V. C. Arturo, Luciano M. Braz (Genesis do Brasil).
The cyclic variation of oil and gas prices, the advance of exploration in hostile environments, such as in ultra-deep waters and drilling through long salt formations, as well as new technologies in the drilled wells geometry are challenges in the upstream of the oil industry.
The lessons learned in the development of such fields are an important source of knowledge that can be used to improve the construction of new wells, fulfilling industry needs regarding economical, environmental and operational levels, based on the E&P constant changes.
One way to use this knowledge is to apply it in new projects, making sure is used in order to improve the technical solutions and mitigate risks, thereby avoiding or minimizing the effects of abnormalities found.
This paper presents a post analysis methodology to implement a structure and data organization of drilled wells in an oil field.
The developed work is within a larger project to combined geomechanical and operational information in order to achieve a holistic view of the field development in years to come.
This research is under way in an offshore field in Brazil and has an objective not only to show the lessons learned, but also to identify the best in class examples to be followed in the future operations of this field.
Dr Kazuo Miura, Marcela Coelho, Petrobras
Dr Carlos Damski, Genesis Petroleum Technologies
Luiz Felipe Coutinho Martins, Magda D C Albuquerque, Maria Alice
Prudente, Mariana D Nascimento and Jéssica L Cordeiro, Genesis do Brasil
A Petrobras Offshore Business Unit has been applying the well-known Deming’s PDCA Cycle (Plan, Do, Check and Act) aiming to reach a continuous optimization in well completion and workover activities (called as intervention in this paper).A methodology based on Ontology of Operations has been implemented to aid this process of continuous improvement.
The methodology defines the intervention planning as a sequence of standardized operations. These standardized operations are the key point of the methodology. We identified more than 180 operations and defined them based on ontology of operations (Miura, 2004). The activity durations are found in a historical database with more than 600 interventions, spliced in conformity with standardized operations. When splicing the interventions in operations, we can obtain larger similarity among the executed tasks than entire interventions, and that allows its statistical treatment.
This same statistical database is useful, for instance, following up performance of an intervention through the simulations which gives us the statistical variation of operational times. For instance, the Monte Carlo’s simulations applied in the operational sequence allow the risk analysis of operational times. Historical data can be queried by year, rig type, oil field, and so on, and in this way, we can get the best statistical distribution curve that fit with the operations times. The analysis that we conducted over these data suggest us that lognormal distribution are the best pattern for these time related operation data.
The virtual intervention simulation technique can be interpreted as a normalization process for operational sequences and it can be used, for example, to benchmark performance of all players involved in workover activities, from the asset team to the service companies.
The methodology allows, in addition, the accomplishment of the intervention post-analysis. The post-analysis makes possible the discovery of operations executed with extreme long or short times, showing both the best durations that should be pursued and the worst durations that should be avoided in future interventions. The abnormal operations are studied in full detail and the lessons learned are incorporate, updating them
in the templates used in the planning of future interventions. With these initiatives, the methodology can assure continuous quality improvement through a positive feed-back.
This paper discusses the implemented methodology and proposes as a new step, the concept expanded to all well operations, including the drilling activities.
Abstract, May 2008
Lessons learned and consequential knowledge acquired during operations are an important base for the design of a new well to be drilled. Shallow hazards, lost circulation zones, abnormal pressures, existence of CO2 or H2S, data obtained during pressure tests (formation and injection tests) that can validate the adopted geopressure model, loss of wells or part of wells are, among others, information that can be gathered from the assessment of drilled wells. The offset wells analysis usually brings a great deal of important and practical results.
In a field offshore Brazil, a Field Knowledge Management model is being built with the aim of developing a methodology to guarantee that the Well Engineering experts know and systematically apply the relevant information collected from past drilled wells in the design of future wells, above the use of offset wells data. The project is intended to produce a set of procedures to allow the use of the model in any oil field.
Following the well-known P-D-C-A (Plan-Do-Check-Act) cycle, the project points to an improvement in quality of data for the “P” phase that comes from organization and verification of data already known from phase “D” of past wells, analysis of these data during phase “C” and finally rearranged in a workflow in phase “A”, this time aiming to facilitate and improve the design of the next well.
As a consequence of this work, experts will have available, in an organized way and with high quality, all the relevant information to support decision process during the elaboration of the next well engineering project. More than 180 wells already drilled in the field will be taken into consideration in this study.
Even more important than Field data analysis will be the methodology under development which will permit self improvements as far as the experts increase their knowledge of the oil field under consideration. This work intends to present the updated development of this project and conclusions already gotten.
Abstract, April 2005
The industry commonly accepts that information from actual drilling, completion and work-over operations should play an important role in the analysis of performance, comparison against plans, optimisation of future operations, and operational safety. It also expects that considerable operational savings would come from the effective use of such information.
In many cases, however, companies cannot experience these gains, even when using sophisticated reporting systems and software tools. Basically this is so because of inappropriate information management processes. For example, sometimes companies do not store planning information in their databases and, thus, cannot compare planned vs. actual. In other cases, even when planning information is kept, it is difficult to match them with actual operations. Most commonly, the operations are not identified or coded properly, making it impractical for the user to perform any reasonable analysis.
This paper describes a new methodology and its practical implementation, which allows for the realisation of all the benefits that reliable data can entail. The basis of the process is the report on planning concept, which changes the current reporting paradigm used by the industry. In this process, the source of operational information shifts from the rig site to the planning office. In addition, the process requires a detailed design of the operational sequences, levels of granularity, and operational code standards.
Abstract 1, September 2004
Drilling performance is usually benchmarked by analysing the time spent on well construction and the degree to which planned times and budgets are achieved. Learn-curve analysis can provide indicators of relative improvement over a campaign, and position on the learning curve can provide a measure of process maturity.
However, the variability of the data makes it difficult to obtain a good fit to the model and assess maturity. The principal cause of the discrepancy between planned and actual times is trouble. Trouble is accounted for in planning by a multiplicative contingency factor. This implies that trouble time is dependent on well or well-section length.
Based on a standardized analysis of drilling performance for a wide range of wells we find that this assumption is not valid and that trouble time is not significantly correlated with well or well-section length. However, we do find that the probability of trouble is significantly correlated with length, and that a probability plot of trouble time for a group of wells or well sections provides a robust measure of drilling process maturity.
We present results of applying this analysis to a variety of well types and propose some metrics for assessing drilling process maturity.
Abstract 2, September 2004
This paper proposes a methodical structure to embody current expertise for the analysis of drilling data. This structure is based on two aspects:
the review of published literature, representative of present-to-day efforts to perform such tasks and
the use of the Methodological Pyramid Concept, to illustrate what elements typify a methodology.
The review of published works revealed that several key aspects of the analysis of drilling data are not clear, for example:
Actual use of drilling data is still unclear and under debate;
Most of the available engineering tools limit the range of applications to drilling performance estimators;
The proposed methods of analysis vary depending upon the needs of individual organizations or processes using the concept of “method and methodology” indistinctly;
Various sets of concept and theories from internal and external sources of drilling engineering knowledge have been used to develop existing methods and engineering tools.
Under this scenario, it seems that more effort is needed to unify current approaches to analyse drilling data i.e. within the scope of an independent field with common goals, theories, methods and tools that can support the decision-making process.
For this purpose, the methodological pyramid was chosen as a convenient model to outline these efforts. By allocating some of the reviewed approaches within the elements of such a pyramid, and contrasting their different ideas where necessary, a Methodology for Drilling Analysis was formalised.
It is believed that this methodology is a convenient framework for defining the goals and scope for an area of drilling expertise named Drilling Analysis and the evolving future role of the Drilling Analyst.
Abstract, October 2002
The study of overall well quality is a neglected area of the industry. Aspects such as formation damage, drilling interactions, and hole rugosity tend to be treated separately in the literature.
This paper describes a software tool that uses stored well data and drilling experiences to produce an overall set of well quality metrics that facilitate the assessment of drilling and completion performance in a given field. The tool is particularly useful for risk assessment when drilling complex trajectories and for improving the design of new deepwater wells.
Abstract, September 2002
Although knowledge is predicted to surpass oil and gas reserves as the most important asset in 21st century oil and service companies, engineers remain cynical about the benefits of current knowledge management initiatives.
This paper discusses problems such as the strong bias of existing knowledge management systems in favour of document searching and focuses on how overlooked areas can be addressed. Results of a collaborative project with 5 major oil companies aimed at overcoming some important limitations of current knowledge systems are discussed.
A key feature of the initiative is software to integrate knowledge capture and reuse into normal work processes. The software uses well data and stored drilling experiences, including problems and solutions, from a global drilling and completions database provided by the member oil companies.
Experiences with deployment of this software into oil companies will be covered as well as the results of research projects on automated learning and case based reasoning to enhance the company knowledge base for optimal design of new wells.
The study of overall well quality has also been included in the project. By defining well quality metrics and then examining them collectively, we can better assess drilling performance in a given field, and ascertain if the company is using its knowledge to improve new well production and cut costs.
Simon Kravis and Rosemary Irrgang, “A Case Based System for Oil and Gas Well Design”, in 15th international conference on Industrial and Engineering applications of Artificial Intelligence and Expert Systems, Cairns June 17-21 2002.
Irrgang, R., Irrgang H., Kravis, S., Irrgang S., Thonhauser, G., Wrightstone A., Nakagawa, E., Agawani, M., Lollback, P., Gabler, T., Maidla, E., “Assessment of Risk and Uncertainty for Field Developments: Integrating Reservoir and Drilling Expertise” SPE Annual Technical Conference and Exhibition, New Orleans, USA, SPE 71419, Oct, (2001).
R Irrgang, S Kravis, P Scott, T Gabler, E Maidla, G Thonhauser, “The Drilling Club: a Knowledge Exchange for Best Practice Operations”, Petrotech New Delhi, Jan 9-11, 2001.
Irrgang, R.; Maidla, E.E., Damski, C.; Millheim, K., “A Case Based System to Cut Drilling Costs”, SPE Annual Technical Conference and Exhibition, October 1999.
The best wells in a basin may be drilled twice as quickly as the worst. New technologies such as slimhole and extended reach drilling are not easily mastered. What factors determined the success of the best wells? What caused the delays? Data management processes and computational tools for planning new wells have been developed to utilise past experiences recorded in drilling data and reports. The Case Based Reasoning technique is applied to derive new scenarios (alternate drilling plans) based on previous “cases”. A number of oil companies are collaborating to form a “Drilling Club” and have contributed well data and drilling reports from around the world. An immense amount of data and experience is available from many wells but engineers would prefer to see just the relevant information. The system assists by extracting an intelligent summary and structuring information to highlight key factors that will result in the greatest cost savings on a proposed drilling operation. Scenarios using alternative technologies can also be tested with risks calculated based on previous similar cases.
Irrgang R. , Kravis S. P. , Agawani M. M. , Maidla, E. , “Automated Extraction of Drilling Experience: Capture and Reuse of Engineering Knowledge”, Petrotech Conference, Jan 1999, New Delhi.