Application of Reliability Engineering to Offshore Production Equipment

ABSTRACT Standard Oil Co. of California performed a reliability study on their subsea completion system in 1973 using an "event-consequence" approach. The study was conducted by a team of company personnel and contract personnel. In this paper, the experiences of company personnel participating in the study are reported by the following format: Definitions of Reliability Study Approaches; Selected Samples for the Purpose of Presenting the Approach Used in the: Reliability Study of a Subsea Completion System; and Selection of the "Optimum Study Approach." INTRODUCTION Although many papers have been written on the different types of reliability studies, few have given concrete examples. This paper shows where the event-consequence approach fits in the range of possible types of reliability studies and then shows by specific example the steps that are required in completing an event consequence study and the types of results that can be obtained, and then, based on experience with the subsea completion system study, a method for choosing the optimum reliability study approach is presented:." Definitions of Reliability Study Approaches One can conveniently separate between three levels of reliability analyses, namely: hazard analysis; traditional reliability analysis; and event-consequence analysis. Hazard Analysis Hazard analysis is hereby being defined as a qualitative evaluation of hazards to the environment of the equipment (system) under consideration. This definition of hazard analysis seems to have become generally accepted as far as the oil industry and its regulatory governmental agencies are concerned. The steps taken in :performing such an analysis are as follows:define the system (equipment) to be considered,define the hazards, such as oil and gas pollution, life safety,determine in a qualitative way what has to fail in order for a particular hazard to take place, andcategorize, the hazards as major, minor or insignificant, depending upon type and/or quantity of hazard, and how many levels of redundant safety systems that have to fail for the hazard to occur. The hazard analysis is in most cases the simplest one to perform. In many ways it comes very close to the inform evaluation a designer makes when he asks himself the question, "What if?", with the exception that the hazard analysis is limited to considering only "damage" to the environment "outside" the system under consideration. Hence, while such an analysis properly conducted could reasonably well protect other interests, it does not give any information on whether the system fulfills its objectives. Traditional Reliability Analysis Traditional reliability analysis applies known failure-rate data in evaluating the probability of a particular event occurring. The steps taken in performing such an analysis are:define the system (equipment) to be considered,select failure events,develop models (failure event description),gather component reliability data (failure rates), andcalculate the probability for occurrence of the event. If the failure events under this form of analysis are the same as the ones under the hazard analysis just described, it would become a hazard analysis except that now the possibility of occurrence has been quantified.