Probabilistic OCTG design: definition, scope, status - interview with ARCO Exploration and Production Technology's M.L. Payne; oil country tubular goods - Interview

World Oil, July, 1998 by Mike Payne

World Oil asked ARCO's Dr. M. L. (Mike) Payne, lead author of the accompanying article on this important industry move toward a more focused tubular design technique, to provide some additional insight into where we are in this evolutionary engineering project, and where we are headed. The following presentation is his response to our questions.

Q. Dr. Payne, as an introduction of this type of OCTG design, what is meant by the term "probabilistic?" Hasn't it also been called "risk-based"?

A. In conventional design, the OCTG performance properties, i.e., burst, collapse, tension ratings, etc., are each defined by one simple number which attempts to describe the minimum performance that tubulars will provide in that loading mode. In probabilistic design, more information is gathered about the pipe, which allows a probabilistic distribution to be developed that more accurately describes what loads and pressure the tubular can withstand.

The same refinement can also be applied to the field loads, i.e., in conventional designs, assumed maximum load cases are used for design, whereas probabilistic design uses refined field loads based more accurately on what can actually occur in that specific field.

The term probabilistic simply means the design process is based on more accurate probability distributions for performance properties and loads rather than the simple and crude assumptions used in conventional designs. Certain implementations of these techniques are also referred to as "risk-based designs."

Q. Basically, what is the designer doing differently than in the "conventional" approach"?

A. The fundamental principle of designing the most cost-effective tubular that safely provides the necessary well integrity and operational functionality is unchanged. What does change is that this design objective is pursued with a much more accurate set of variables to characterize the performance properties for the tubular and the field loads which must be sustained. In many ways, the approach boils down to taking a "sharper pencil" to conventional design practices.

Q. Can you describe what is needed for this new approach to be utilized?

A. There are two key areas which enable use of this new design technology:

1. The ability to develop comprehensive information on the pipe performance properties by looking at the pipe's dimensional and mechanical data, or by directly testing the performance, and

2. The ability to develop or obtain comprehensive information on field loading.

For performance properties, the operator could independently re-inspect the pipe to obtain data on wall thickness, yield strengths, etc., but this can be expensive and time-consuming. A more recommended approach is for the operator to work closely with the pipe manufacturer or supplier to obtain the data necessary to describe the actual performance properties. This can be as simple as agreeing to make the pipe with a minimum wall thickness better than API's minimum of 87.5%, or it can be more comprehensive.

A more comprehensive implementation would include statistical reports on pipe wall thickness, ovality, yield strength, ultimate strength, toughness, etc. These statistics are used to calculate enhanced performance properties for the pipe. Additional measures can include full-scale collapse testing or use of elevated hydrostatic (internal pressure) testing.

Developing comprehensive information on field loading requires that the operator study his operational files and possibly those of offset operators. Through such studies, a better understanding of realistic field loads can be developed and excessive overdesign can be eliminated.

Q. How would collapse testing and elevated hydrostatic testing help?

A. API's collapse ratings are based on statistical analyses of fairly old databases of pipe samples. Some of the collapse data goes back to the early 1960s. For the most part, current manufacturers can now make pipe much better than what was produced in years past. Being able to quantify how much better your pipe is than the old pipe used for the API collapse equations can be very leveraging in allowing more cost-effective designs.

Our work has shown some margins as high as 20-50% between actual performance and API ratings. It should also be noted, however, that we have found some manufacturers whose pipe is basically in line with API performance properties and not any better. Being able to quantify the differences between the many manufacturers now in the marketplace is critical. For burst-driven designs, elevated hydrostatic testing provides a cost-effective, fast and safe way of demonstrating fitness for purpose for pressures up to, and even slightly exceeding, the API MIYP pressure.

Q. Does this approach offer advantages for certain types of operations, i.e., is it more applicable to severe applications? Would conventional methods be OK for a 7,000-ft straight hole in an established area such as West Texas?

A. The principles of these techniques are universal. We have successfully applied them on expensive HPHT exploration wells in the South China Sea and on shallow, vertical development wells in West Texas. The per-well cost saving is obviously larger on the more expensive wells, but if you are going to drill a lot of development wells in West Texas, or elsewhere, several thousands or tens of thousands of dollars savings per well can add up to several hundreds of thousands or even millions of dollars over time. Because of this, the techniques have very large potential across the industry.