PROBLEM TO BE SOLVED: To provide a method for generating an integrated evaluation model for various phenomena at the time of an earthquake, which can evaluate and predict various indexes of various phenomena at the time of an earthquake caused by an earthquake in an integrated manner and with high accuracy. SOLUTION: The method for generating an integrated evaluation model for various phenomena during an earthquake is described in a first database 10A to store the actual earthquake characteristic parameters and the index observation results when various phenomena due to an actual earthquake are observed. The first acquisition step of acquiring a plurality of learning data 12A of 1 and the second acquisition of virtual earthquake characteristic parameters and index calculation results when various phenomena at the time of an earthquake due to a virtual earthquake are calculated by simulation are stored in association with each other. As a trained model of machine learning, based on the second acquisition process of acquiring a plurality of second learning data 12B from the database 10B, and the plurality of first learning data 12A and the second learning data 12B. It includes a generation step of generating an integrated evaluation model 13 for various phenomena during an earthquake. [Selection diagram] Fig. 3
展开▼