Drilling industry benefits by sharing successful drilling practices

Mark K. Gregoli, J. Ford Brett
OGCI Management Inc.
Tulsa

Brian C. Gahan
Gas Research Institute
Chicago

Benefit can be created when drilling organizations systematically cooperate to identify and share successful practices.

The successful drilling practices (SDP) approach identifies, documents, disseminates, and measures the benefits for a given study area. This method provides operators with performance benchmarks and an ongoing knowledge base.

As this article will show, successful drilling practices consist of aggregated, small-scale tasks that serve to improve the overall drilling process. It is the summation of these incremental improvements that accounts for the majority of cost reductions and improvements in drilling efficiencies, as opposed to larger-scale, single-process improvements. The SDP approach strives to capture and share these small-operating practices with the industry.

When operators wish to identify and document such practices, they can be hindered by a general lack of detailed operating records. Morning reports fail to document all the "small tricks of the trade," and an operator may lose experience with personnel turnover.

Furthermore, without the consent of other operators, it is impossible for any single operator to collect and analyze the detailed drilling information required to identify successful practices. Only through joint-operator efforts can the successful practices be compiled for industry-wide use.

Because of the "remembering" problem, a systematic catalog of successful practices will most likely yield broad-based benefits.

SDP studies

Eleven separate SDP studies funded by the Gas Research Institute (GRI) were commissioned to:

• Identify the most successful drilling practices for each of the study areas.
• Document these practices in enough detail to ensure practical application.
• Disseminate the successful practices.
• Measure the benefit derived from these studies.

The studies, completed between 1994 and 1998, focused on the following drilling environments encompassing 10 areas:

1. Greater Green River basin-Two studies: South Moxa Arch Dakota formation and Wamsutter Almond formation
2. Arkoma basin-Choctaw thrust, Spiro formation
3. Arkoma basin-Jackfork Play
4. Anadarko basin-Watonga-Chickasha trend, Morrow/Springer formations
5. Wilcox Lobo trend, Webb County, Texas
6. Deepwater exploration drilling, Green Canyon area, Gulf of Mexico
7. Extended reach drilling, South Pass area, Gulf of Mexico
8. Deep Austin Chalk, Washington County, Texas
9. Cotton Valley reef play, East Texas
10. Val Verde basin-Strawn and Penn Sands, Texas.

Studies have shown that drilling costs can vary by as much as 50% for wells drilled in very similar geologic environments with identical objectives. Figs. 1-4 further support the existence of variability in drilling performance under what appears to be seemingly similar conditions.

The figures show every well drilled over a period of years in four different areas. Even under these identical conditions, the figures show a high degree of variability, on the order of 50%. There are two probable causes for producing this variability:

1. Differences in the specific geological formations encountered while drilling
2. Differences in controllable drilling parameters.

It can be reasoned that the source of this variation was caused by random differences in the geologic sequence such as slightly harder formations, differences in hole stability, and fault zones.

At face value, the figures might be used to support such a claim. However, Fig. 1 [36,618 bytes]Fig. 2 [46,654 bytes]Fig. 3 [38,360 bytes]Fig. 4 [40,866 bytes] show performance improvements in the range of 5 to 15% each year for several years running. This improved performance means that the observed variability was not caused by random and uncontrollable differences in the geologic sequence, but instead, was caused by differences in the application of technology and procedures.

On any given well, the worst performing operations might rationalize that they were just "unlucky" and happened to hit the more difficult areas of drilling. The figures, however, show that even the worst performers improve with time.

These data imply that the performance can be, and has been, controlled. Operators have proven they can reliably reduce drilling time by systematically changing operational procedures and processes.

The statistical evidence for this improvement is shown in Figs. 1 and 2. These figures show a linear regression with a coefficient of determination (R2) of 0.35 and 0.22, respectively. That means the regression line accounts for 35% and 22% of the variability in the data sets.

The P-values (probability) of the T-Test (test statistic), however, show that even though the regression line describes a small percentage of the variability, there is less than a 1 in 1,000 chance that the improvement in performance shown is due to statistical chance. The huge variation is controllable, as operators are able to improve performance with time.

The high variability in drilling performance can be explained by looking at a drilling operation as an imperfect application of experience. Operators drilling only a few wells in a particular area have a greater than 74% chance of performing below average (Fig. 5 [38,838 bytes]).

Operators involved with limited drilling programs, however, do not account for all of the below-average wells. Clearly other operators find it difficult to retain the benefits of experience caused by personnel turnover, incomplete documentation, and time lag between operations.

Information capture

Another potential reason for the large variability is the absence of a systematic way to capture and disseminate drilling experience within a company.

Table 1 [62,310 bytes], which is based on the same set of data used to develop Figs. 1-4, summarizes the differences between the average well cost and the most successful well cost. If it were possible for the average well to reliably achieve successful performance, there would be a significant decrease in well costs.

A basic principle of quality management is that variation in performance means opportunity for improvement. The data shown in Figs. 1-4 support this principle. The data also imply that in almost every circumstance, no single operator has a "lock" on the most successful practices.

In only one of the 11 areas did a single operator have a clear advantage (Fig. 3). This means that it is uncommon for one operator to consistently outperform the industry, and that every operator has the potential to learn from others.

Controlled variability in drilling performance is economically significant-encompassing about 20% of well costs. Thus, the industry can benefit by improving the way it gathers, shares, and disseminates successful drilling practices.

Current experience

The primary formal cooperative mechanism for capturing and sharing successful drilling practices consists of operator interaction with drilling contractor and service-company technical-sales organizations. These organizations have been very quick to share results with potential customers.

Because such organizations have a profit incentive, they can be counted on to share successful practices, performing a service to the industry by doing so. There are many examples of successful practices being cooperatively developed in this way, including technological advancements in horizontal drilling, top drives, synthetic muds, and polycrystalline diamond compact (PDC) bits.

As effective as technical sales organizations can be, there are some difficulties with this mechanism. First, corporate self-interest can get in the way of broad dissemination of best practices. Each drilling contractor and service company is going to make the best possible case for its own particular brand of technology.

Although this free-market approach provides the customer with choices, it sometimes slows the transfer of improved technology and keeps inefficient techniques around longer than warranted. Second, improving overall industry efficiency is not the focus of these organizations; instead, corporate profit comes first.

An example of such a situation occurred early in the history of PDC bits when manufacturers had no vested interest in eliminating four or five tri-cone bit runs with a single PDC bit.

In response to competitive pressures, the market drives organizations to improve performance. Although technical sales organizations can efficiently, quickly, and widely distribute large amounts of detailed, technical know-how, in many situations they are structurally prevented from promoting the global optimum.

A second approach to technology transfer comes through communications between formal and informal networks of technical professionals. Examples of informal networks are rig crews "swapping war stories" and drilling managers discussing work while playing golf. Examples of formal networks are side discussions that occur during Society of Petroleum Engineers (SPE), American Association of Drilling Engineers (AADE), and American Petroleum Institute (API) meetings.

Formal and informal networks can be effective at identifying common problems, scoping solutions, and exchanging successful practices. Sharing experience in this way is easy, and often quite enjoyable for those involved.

However, these approaches can suffer from the disadvantages of not providing a reliable way to identify truly successful practices. For example, can it be reliably ascertained that the person who provided the proposed solution actually solved the particular problem at hand?

Such discussions also lack technical depth and can be limited in the amount of real technical material that is developed and shared. In addition, such information is normally distributed to a limited number of people.

The formal and informal networks can identify and disseminate successful practices. But these networks do not reliably ensure that the most successful practices are developed or shared. In addition, the speed by which they transmit information is limited by the size of the network.

Finally, formal and informal networks do not always effectively facilitate information exchange, which by necessity must be very detailed. This last point is particularly important, especially since the value of successful practices is often found in small details.

Another mechanism of technology transfer occurs through the publication of technical papers and journal articles. Articles have the purpose of describing a common concern and sharing an approach to address the concern. Unfortunately, this approach also has certain limits.

First, technical papers and journals are normally published long after the fact and are therefore not very timely. Second, the existence of technical publications does not necessarily guarantee the sharing of the most successful practices. Because they take quite some time and effort to write, and provide few direct financial incentives to the author, technical forums cannot be relied upon to document and disseminate successful practices.

Third, restrictions on publication length often prevent the reader from receiving enough detail to reproduce the successful practices. Technical papers often deal with large or unique technical issues, not mundane operating practices.

One mechanism that can quickly transfer extremely detailed technical know-how is the movement of personnel between organizations. This approach disseminates successful practices throughout the industry and does not necessarily have to include predatory hiring.

Finally, there are examples where the industry has chartered cooperative efforts, such as API standards committees, to investigate and develop solutions for mutual concerns. The recent Mobil-BP-Texaco-Chevron (MoBPTeCh) cooperative is another example.

Because pooled resources serve to attack common problems, cooperative efforts can provide cost-effective, highly detailed solutions. Unfortunately, because they often take a great amount of time and effort to organize, they have been traditionally limited to specific technical issues and the bigger challenges that face the industry.

Cooperative benefits

In mature areas, the drilling business has many striking parallels with the farming business, an industry that has a strong cooperative system. In such markets, no single operator can possibly affect the price of oil or natural gas, maintain the support infrastructure, or efficiently assess every possible technique. Like farmers, solving common problems and sharing successful practices benefits all without high individual cost.

To answer the question of cooperative benefits and to achieve the goals listed at the beginning of this article, the following technical approach has been developed.

Study areas were selected based on drilling activity, operator interest, and an internal GRI analysis of possible resource bases. Readers are referred to the studies in the bibliography for a complete description of the study methodology.

Once an area was selected, the following six steps were used to produce a detailed drilling operations plan for a successful well:

1. Build an initial data base.
2. Solicit operator participation and collect detailed drilling reports.
3. Conduct preliminary analysis of drilling reports.
4. Select prototype well.
5. Conduct in-depth interviews.
6. Prepare drilling operations plan.

A 1-day SDP workshop is conducted for each study area. The workshops consist of the presentation of each study's key findings, presentations by operators and service company personnel associated with the successful operations, and round table discussions.

Manuals produced for the studies are distributed to all operators, contractors, and service company personnel providing information (available from the GRI).

Independent verification

To validate the SDP approach, Gelb Consulting Group Inc. conducted an independent survey consisting of 87 telephone surveys and 42 in-depth interviews. The interview candidates were randomly selected from a pool of those who attended a workshop or received a report, who claimed their organization realized a benefit, and who reported that they spent more than 70% of their time on drilling-related planning or operations.

The interviewees represented a reasonable cross section of the industry and job classifications (Table 2 [34,644 bytes]). More than half of the interviewees cited a specific change in drilling or business practices attributed directly to the SDP reports and workshops. Dollar savings per interviewee ranged from $50,000 to $1.3 million.

Respondents who were unable to provide a direct measure of financial impact were asked to estimate the percent savings that could be realized through implementation of the best practices. Seventy-five percent thought savings would fall in the 5-10% range, 17% estimated potential savings greater than 10%, and 8% estimated savings of less than 5%.

One important consideration is the required frequency of study updates. Because the studies represent a snapshot of successful practices at only one point in time, they may become outdated as technology and practices improve. In response to a question inviting a suggested update, 64% reported that the studies should be updated at least every 2 years.

Greater Green River basin

Fig. 6 [57,912 bytes] presents statistical evidence suggesting that successful practices can be identified and transferred as described above. The figure shows an industry-wide decline of 10% in the number of days it took to drill 10,000 ft in the Moxa Arch area from 1991 through 1993 (from 26.2 to 22.7 days/10,000 ft).

In 1994, one operator successfully instituted a number of technical and business process improvements on wells in the same area and achieved similar or better performance. As shown in the figure, these improvements resulted in a 41% step-change reduction in drilling time to an average of 13.5 days/10,000 ft.

The cause of this step change became the subject of two successful practice studies. Of interest here are the measurements of the effectiveness of this approach as a means for transferring this significant improvement to the area in general.

In early 1996, the 1995 drilling results for area operators was compared with those who did and did not participate in the SDP studies. Fig. 6 shows that the expected disparity actually occurred with study participants out performing those who chose not to participate.

The operator that made the 1994 step change achieved similar performance (13.5 days) in 1995. Most interesting, Fig. 6 also shows that study/workshop participants reduced average drilling time by 33% (22.7 to 15.1 days/10,000 ft), while study/workshop nonparticipants improved performance by only 13% (22.7 to 19.9 days/10,000 ft).

Although the 33% improvement experienced by participants is less than the 41% of the most successful operator, the improvement of the nonparticipants was on trend with the regular 4-year average.

A statistical T-Test was used to determine the probability of similar performance differences between the successful operator, study/workshop participants, and nonstudy/workshop participants.

The test showed that in 1995, participants had a 44% probability of repeating the performance of the highly successful operator; while nonparticipants had only a 4% probability of equaling the successful operator's results.

The point is that the improvement shown by the participants was almost certainly not the result of chance but the result of an intentional design to cooperatively identify, share, and implement successful practices.

Bibliography

1. Brett, J.F., Milheim, K.K., "The Drilling Performance Curve: A Yardstick for Judging Drilling Performance," SPE paper 15362, Annual Technical Conference and Exhibition, New Orleans, 1986.
2. SDP-Anadarko Basin: GRI-95/0132.3, Gas Research Institute, Chicago, 1996.
3. SDP-Arkoma Basin: GRI-95/0132.2, Gas Research Institute, Chicago, 1995.
4. SDP-Deepwater Exploration Drilling: GRI-95/0132.4, Gas Research Institute, Chicago, 1996.
5. SDP-Extended Reach: GRI-95/0132.6, Gas Research Institute, Chicago, 1996.
6. SDP-Greater Green River Basin: GRI-95/0132.1, Gas Research Institute, Chicago, 1995.
7. SDP-Wilcox Lobo Trend: GRI-95/0132.5, Gas Research Institute, Chicago, 1996.
8. SDP-Technical Process Integration: GRI-96/0232, Gas Research Institute, Chicago, 1996.
9. Successful Practices Drilling Operations: Valverde Basin, GRI-97/0400.4, Gas Research Institute, Chicago, 1998.
10. Successful Practices Drilling Operations: Deep Austin Chalk, GRI-97/0400.1, Gas Research Institute, Chicago, 1998.
11. 11. Successful Practices Drilling Operations: Arkoma Basin, Jackfork Play, GRI-97/0400.3, Gas Research Institute, Chicago, 1998.
12. 12. Successful Practices Drilling Operations: Cotton Valley Reef, GRI-97/0400.2, Gas Research Institute, Chicago, 1998.
13. 13. Gregoli, M.K., Brett, J.F., Gahan, B.C., "The SDP Study," SPE paper 38636, presented at the SPE Annual Technical Conference and Exhibition, San Antonio, 1997.
14. 14. Reeves, S.R., "Benefits of SDP: Greater Green River and Arkoma Basins," Final Report, March 1995.

The Authors