Uncertainty treatment in production assurance analyses throughout the various phases of a project

In a production assurance analysis, we study the performance of production and transportation systems in relation to the demand for deliveries. Different measures can be used to express the performance of the systems, and there are different ways of presenting the results. The different performance measures reflect to varying degree uncertainties, as a result of variation and lack of knowledge, and it is a challenge to establish a proper structure for dealing with the uncertainties. The purpose of this paper is to discuss and give guidance on these issues, emphasizing the treatment of uncertainties in the production assurance analyses throughout the various phases of a project, prior to production. We recommend the implementation of a Bayesian approach based on prediction intervals, with different characteristics dependent on the project phase.