Development of bias in analytical predictions based on behavior of platforms during hurricanes
A Joint Industry Project (JIP) was initiated by 13 oil companies and the US Minerals Management Service (MMS), wherein a methodology was developed to use information from observed platform conditions resulting from Andrew and the hurricane hindcast data with capacity, reliability, and Bayesian updating analyses to determine a measure of differences (biases) in the analytical predictions and field observations. The procedures used for structural integrity analysis were also improved as a result of this study. Phase 1 of this project completed in October 1993 defined a global bias factor. A study of foundation behavior was completed following Phase 1 and determined bias factors specific to foundation failure modes. This paper presents the approach followed in the most recent phase of this project in which bias factors specific to jacket and two foundation failure modes (lateral and axial) were developed. This study utilized an updated storm hindcast, improved analysis models, and a more detailed calibration procedure. The three bias factors were developed and were found to differ significantly. The bias factors developed through this study have provided means to further improve procedures used in the assessment of existing platforms. The proper use of these new analytical methodologies and bias factors will produce more appropriate and cost-effective mitigation measures for safe platform operations. The methodology for establishing bias factors developed and proven in these projects is applicable to other offshore regions and production systems with specific environmental, geotechnical, material and structure features.
- OSTI ID:
- 433936
- Report Number(s):
- CONF-960525--
- Country of Publication:
- United States
- Language:
- English
Similar Records
Results from large scale ultimate strength tests of K-braced jacket frame structures
Capacities of template-type platforms in the Gulf of Mexico during hurricane Andrew