Periodical inspection is necessary for stable operation of the plant. But usually, inspection cost and inspection period are limited, and it is difficult to inspect all of the objective location of the plant in allowed cost and period. Therefore, for maintenance planning, probability of failure is presumes by the obtained sample size inspection data. In that case, presumed probability of failure includes uncertainty, and the one-side tolerance limit of the probability is changed by the sample size of the inspection.
For maintenance on boiler tubes of fossil fuel fired power plants, only limited sample size inspection data can be used for maintenance planning. On the other hand, the conditions of the tubes for maintenance are different because environmental conditions of the tubes like actual temperature are different. Then the inspection results don’t demand enough information to presume probability of failure because the standard deviation is not known. But usually to presume probability of failure for maintenance, the standard deviation is not considered only the average is used.
Therefore, considering mainly uncertainties depend on sample size of the inspection data, presumed probability of failure is related to rational maintenance planning.
