Using experts’ opinion for Bayesian prior reliability distribution of on-demand equipment: A case study of a novel sliding sleeve valve for open-hole wells

Abstract

Reliability estimation is crucial to support analysts to decide whether it is feasible to proceed with the equipment development. A problem arises in the early stages of equipment development, where reliability data is often scarce. Bayesian inference is widely applied to estimate technologies’ reliability over their lifetime as it allows updating prior knowledge as field and/or test data are gathered. This study proposes a methodology to build informative prior distribution for different failure mechanisms of novel equipment operating in standby mode and working on demand. The methodology is flexible enough to consider events that degrade over time and/or with past demands. In addition, it involves an elicitation process to extract experts’ opinions on the failure probability at the system level. The uncertainty regarding the different experts’ opinions is considered by modeling each expert's opinion with a probability distribution and then aggregating them via multiplicative error and Population Variability Analysis. Thus, the proposed methodology considers how probability estimates vary among experts and defines a Bayesian informative prior distribution using only experts’ knowledge. Finally, we present a case study involving a novel sliding sleeve valve of large diameter that will operate in open-hole wells of the Oil & Gas industry.