Probabilistic Scheduling of Cavitation Repairs

See related content

by James E. Giles, TVA Engineering Lab, Norris, United States,
Patrick A. March, TVA Engineering Lab, Norris, United States,
Walter O. Wunderlich, TVA Engineering Lab, Norris, United States,

Document Type: Proceeding Paper

Part of: Waterpower '91: A New View of Hydro Resources

Abstract: The probabilistic nature of maintenance scheduling is exemplified by operating histories and conditions which vary from unit to unit. Typically, an operator must decide whether to repair a unit that is in a deteriorated condition or to continue operating the unit. The cost of repairing the unit now is partially offset by more efficient and reliable operation in the future. Continuing to operate the unit gives immediate returns that are partially offset by less efficient operation, less reliable operation, decreased equipment life, and increased maintenance costs in the future. Past experience with the maintenance of individual units and judgements by plant personnel can be used to estimate the likelihood of a unit's changing from a particular status, such as 'fully operational,' to another status, such as 'slightly deteriorated' or 'severely deteriorated.' These likelihoods, which are called transition probabilities, can be combined with economic data to quantify operating benefits and costs. This information can be used to evaluate explicitly the expected cost associated with maintenance alternatives. Examples are provided which show how a spreadsheet format can be used to evaluate a decision of whether or not to repair cavitation damage to a hydro unit.

Subject Headings: Maintenance | Probability | Scheduling | Deterioration | Cavitation | Hydro power | History | Equipment and machinery |

Services: Buy this book/Buy this article


Return to search