Model generation apparatus, model generation method, and non-transitory storage medium
A model generation apparatus ( 2000 ) acquires component failure data in which a usage status is associated with a failure record of a component. The model generation apparatus ( 2000 ) generates, for each of a plurality of component groups, a prediction model for predicting the number of failures of each component included in the component group by using the component failure data relating to the component belonging to the component group. The prediction model computes a prediction value of the total number of failures of the components belonging to a corresponding component group from the usage status, and computes a prediction value of the number of failures of each component belonging to the component group from the computed prediction value of the total number of failures.
1. A model generation apparatus comprising:
at least one memory configured to store one or more instructions; and
at least one processor configured to execute the one or more instructions to:
acquire component failure data in which a usage status is associated with a failure record of a component; and
generate, for each of a plurality of component groups a prediction model for predicting a number of failures of each component belonging to the component group by using the component failure data relating to the component belonging to the component group,
wherein the prediction model computes a prediction value of a total number of failures of the components belonging to the corresponding component group from the usage status, and computes a prediction value of the number of failures of each component belonging to the component group from the prediction value of the computed total number of failures.
2. The model generation apparatus according to claim 1 ,
wherein the prediction model comprises a first sub model that predicts the total number of failures of the components belonging to the component group from the usage status, and
wherein the processor is further configured to execute the one or more instructions to generate for each of the plurality of component groups, the first sub model by estimating a parameter of the first sub model by using the component failure data relating to each component belonging to the component group.
3. The model generation apparatus according to claim 2 , wherein
the prediction model comprises a second sub model that predicts the number of failures of each component belonging to the corresponding component group from the total number of failures predicted by the first sub model, and
the second sub model computes the prediction value of the number of failures of each component by using the predicted total number of failures and a predicted distribution which is a distribution of the number of failures or a failure rate of each component.
4. The model generation apparatus according to claim 1 , wherein
the plurality of component groups are generated by performing classification on a plurality of components using domain knowledge relating to a relationship between the components.
5. The model generation apparatus according to claim 4 , wherein
the domain knowledge relates to a correlation between the respective components based on a geometric structure of arrangement or a correlation between the respective components based on an operation status.
6. A model generation method executed by a computer, the method comprising:
acquiring component failure data in which a usage status is associated with a failure record of a component; and
generating, for each of a plurality of component groups, a prediction model for predicting a number of failures of each component belonging to the component group by using the component failure data relating to the component belonging to the component group,
wherein the prediction model computes a prediction value of a total number of failures of the components belonging to the corresponding component group from the usage status, and computes a prediction value of the number of failures of each component belonging to the component group from the prediction value of the computed total number of failures.
7. The model generation method according to claim 6 , wherein
the prediction model comprises a first sub model that predicts the total number of failures of the components belonging to the component group from the usage status, and
the computer generates, for each of the plurality of component groups, the first sub model by estimating a parameter of the first sub model by using the component failure data relating to each component belonging to the component group.
8. The model generation method according to claim 7 , wherein
the prediction model comprises a second sub model that predicts the number of failures of each component belonging to the corresponding component group from the total number of failures predicted by the first sub model, and
the second sub model computes the prediction value of the number of failures of each component by using the predicted total number of failures and a predicted distribution which is a distribution of the number of failures or a failure rate of each component.
9. The model generation method according to claim 6 , wherein
the plurality of component groups are generated by performing classification on a plurality of components using domain knowledge relating to a relationship between the components.
10. The model generation method according to claim 9 , wherein
the domain knowledge relates to a correlation between the respective components based on a geometric structure of arrangement or a correlation between the respective components based on an operation status.
11. A non-transitory storage medium storing a program that causes a computer to execute each step of the model generation method according to claim 6 .