Abstract
Successful management of a population of Electric Submersible Pumps (ESPs) requires an accurate estimate of the expected number of failures as a function of time plus a determination of the associated uncertainty. Knowledge of the uncertainty is required for an effective risk assessment to be made of workover strategies and the impact of these strategies on the production economics. Though the oilfield literature contains discussions of the models and methods for estimating the expected number of ESP failures, little discussion has been presented on the confidence of such estimates.
This paper contains details of how the choice of model, parameter estimates and variation in a well's economic criteria affect the confidence limits. It shows that the uncertainties inherent in the continuous replacement of ESPs and the estimation of their run lives may make the estimates of the number of failures across an entire field to be numbers with considerable uncertainty. This effect is particularly pronounced during the first few years after field start-up. Under these circumstances the distribution of the number of failures is skewed and a 60% difference between the estimated and recorded number of failures is possible. Statistical models also enable differentiation in performance between manufacturers and technologies. It is shown that even large differences in measured run life may not be conclusive evidence of technical or manufacturing superiority. Based on these results, actions are suggested to reduce the uncertainty and improve the management of a field.
