Model utilization system, model utilization method, and computer program product

A model utilization method comprising:

• performing, by a model utilization system, processing of a distributed ledger configured to store a scenario to be used for analysis of a target system, a digital twin model to be used for the analysis, and data to be input to the digital twin model;
• calculating, by the model utilization system, a token that indicates an incentive for an organization that has provided the digital twin model to be utilized and for an organization that has provided the data to be utilized;
• and
• retrieving, by the model utilization system, from the distributed ledger, a candidate of the scenario that satisfies a user request, a candidate of the digital twin model that satisfies the user request, and a candidate of the data that satisfies the user request, presenting the candidate of the scenario retrieved, the candidate of the digital twin model retrieved, and the candidate of the data retrieved, and presenting a token required to use a combination of the candidate of the digital twin model retrieved and the candidate of the data retrieved, wherein
• the target system includes an infrastructure system,
• the data includes at least one of a usage history of the infrastructure system, a load history of the infrastructure system, and monitoring data of the infrastructure system, and
• the digital twin model includes a surrogate model to be used for at least one of diagnosis of an abnormal sign of the infrastructure system, life prediction of the infrastructure system, and failure risk prediction of the infrastructure system;
• and further comprising:
• calculating, based on the candidate of the scenario, a risk calculated from the candidate of the digital twin model and a value indicating usefulness of the candidate of the digital twin model;
• retrieving the candidate of the digital twin model for which the risk and the value satisfy the user request;
• calculating an indicator expressed in a probability distribution by performing simulation using the candidate of the digital twin model and the candidate of the data, based on the candidate of the scenario;
• calculating model reliability of the digital twin model by uncertainty evaluation based on the indicator;
• and
• calculating the risk to be higher as the model reliability is lower;
• and
• wherein the surrogate model includes a model generated from a function candidate library based on the user request.

A computer program product comprising a non-transitory computer-readable medium including programmed instructions, the instructions causing a computer to execute:

• performing processing of a distributed ledger configured to store a scenario to be used for analysis of a target system, a digital twin model to be used for the analysis, and data to be input to the digital twin model;
• calculating a token that indicates an incentive for an organization that has provided the digital twin model to be utilized and for an organization that has provided the data to be utilized;
• and
• retrieving, from the distributed ledger, a candidate of the scenario that satisfies a user request, a candidate of the digital twin model that satisfies the user request, and a candidate of the data that satisfies the user request, and presenting the candidate of the scenario retrieved, the candidate of the digital twin model retrieved, and the candidate of the data retrieved, and presenting a token required to use a combination of the candidate of the digital twin model retrieved and the candidate of the data retrieved, wherein
• the target system includes an infrastructure system,
• the data includes at least one of a usage history of the infrastructure system, a load history of the infrastructure system, and monitoring data of the infrastructure system, and
• the digital twin model includes a surrogate model to be used for at least one of diagnosis of an abnormal sign of the infrastructure system, life prediction of the infrastructure system, and failure risk prediction of the infrastructure system;
• and further causing the computer to execute:
• calculating, based on the candidate of the scenario, a risk calculated from the candidate of the digital twin model and a value indicating usefulness of the candidate of the digital twin model;
• retrieving the candidate of the digital twin model for which the risk and the value satisfy the user request;
• calculating an indicator expressed in a probability distribution by performing simulation using the candidate of the digital twin model and the candidate of the data, based on the candidate of the scenario;
• calculating model reliability of the digital twin model by uncertainty evaluation based on the indicator;
• and
• calculating the risk to be higher as the model reliability is lower;
• and
• wherein the surrogate model includes a model generated from a function candidate library based on the user request.