A Comparison of Deterministic and Reliability-Based Design Methodologies for Production Tubing

Abstract

TX 75083-3836, U.S.A., fax 01-972-952-9435. AbstractIn this paper, we compare reliability-based procedures for design of production tubing with the traditional working-stress design method. The basic approach of Working Stress Design (WSD), Load and Resistance Factor Design (LRFD) and Quantitative Risk Assessment (QRA) are discussed. Load and resistance methodologies for the three approaches, and their impact on the tubing design are contrasted. The relationship between the Safety Factor (SF) used in WSD, the Load Factors (Lf) and Resistance Factors (Rf) used in LRFD, and uncertainity propragation used in QRA is illustrated. These relationships are then compared and contrasted with the determined reliability for a given design. A few illustrative examples are provided for the three different design methods. Graphs illustrating the relationship between SF and estimated design reliability allow a reliability-based interpretation of WSD, and give insight into the design reliabilities implicit in WSD. By means of such a comparison to a familiar design format, this paper aims to give the reader an understanding of, and confidence in LRFD/QRA format for the design of production tubing. The tendency of WSD to overdesign tubing from a reliability stand-point is illustrated. It is concluded that while QRA is the best approach to designing production tubing to a desired reliability, LRFD allows reliability-based design in a format that is relatively easy to apply.