| 摘要 |
<P>PROBLEM TO BE SOLVED: To provide a maintenance planning method capable of acquiring the probability in the vicinity of an important state for planning maintenance, such as whether or not an inspection is necessary, more precisely than before and without requiring massive arithmetic processing. <P>SOLUTION: A weighted probability density function w that has a weight on a damage state in need of an inspection and a repair is determined based on validated data related with a damage state of components or a unit constituting a plant or a product. A uniform random number or the like is set as a cumulative probability of a cumulative distribution function weighted probability density function w, and a probability variable z<SB>n</SB>corresponding to the uniform random number is determined. A weighted probability of the damage state of the components or the unit constituting the plant or the product is calculated from data relating to the probability of the damage state using the z<SB>n</SB>as a random value by Monte Carlo simulation. A probability of the damage state of the components or the unit constituting the plant or the product is calculated from a restoration value W and the weighted probability of the damage state, and based on the probability, the unit or components to be inspected or repaired are selected by calculating the damage state of the components and the unit. <P>COPYRIGHT: (C)2008,JPO&INPIT |
