Abstract
The timing of replacement of equipment is dependent on factors like replacement costs, discount factors and differences in productivity, reliability and safety of existing and new equipment. Predictions related to these factors are, however, normally subject to uncertainties, and in this paper we discuss how two subjective approaches to representing uncertainties related to future events—the classical Bayesian approach and the fully subjective approach—can be used as a basis for decision-making about replacement. Some key features of the approaches are discussed, including integration of engineering judgements, uncertainty treatment and type of performance measures to be used. In the fully subjective approach, we advocate the direct subjective assessment of uncertainty in a quantity of interest, which should be observable, with decisions made on the basis of this uncertainty. The fully subjective approach is illustrated using an example in capital equipment replacement.
Keywords:
Acknowledgements
The work described in this paper was carried out while the first named author was on a study exchange at the University of Salford. We thank the Norwegian Education Office and University College Stavanger for supporting this exchange. We are grateful to Terje Aven for his continuing support throughout, and also for the helpful comments of a referee that significantly improved the clarity of the paper.