Prediction apparatus, prediction method, and program

A prediction apparatus for predicting a remaining capacity value of a battery of a vehicle, the prediction apparatus comprising:

• a processor;
• and
• a memory storing instructions that cause the processor to execute a process, the process including
• acquiring a plurality of pieces of usage record data including information regarding a travel distance of the vehicle and information regarding charging including information indicating a charge amount of the battery or information indicating a charge time of the battery as use records of the vehicle in the past or not including the information regarding charging as use records of the vehicle in the past, and acquiring usage schedule data including information regarding a travel distance of the vehicle as a usage schedule of the vehicle in the future;
• extracting, from among the plurality of pieces of usage record data, specific usage record data including information regarding a travel distance that is the same as or similar to the information regarding the travel distance included in the usage schedule data, and determining that the battery has been charged in a case where the specific usage record data includes the information regarding charging;
• calculating a correction value as a predetermined charge amount based on information indicating the charge amount or information indicating the charging time among the information regarding charging included in the specific usage record data in response to determining that charging has been performed;
• predicting a remaining capacity value of the battery at the time of return with respect to the acquired usage schedule data using a trained machine learning model that outputs a remaining capacity value of the battery at the time of return of the vehicle with the usage schedule data as an input;
• using the calculated correction value to execute correction on the predicted remaining capacity value of the battery at the time of return;
• and
• outputting a remaining capacity value of the battery after executing the correction.

A prediction method executed by a prediction apparatus for predicting a remaining capacity value of a battery of a vehicle, the prediction method comprising:

• acquiring usage record data including information regarding a travel distance of the vehicle and information indicating a charge amount of the battery or information indicating a charging time of the battery as a past use record of the vehicle, and acquiring usage schedule data including information regarding a travel distance of the vehicle as a future usage schedule of the vehicle;
• determining that the battery is charged in a case where a predetermined traveling distance based on the information regarding the traveling distance included in the usage record data is higher than a threshold;
• calculating a correction value as a predetermined charge amount based on information indicating the charge amount or information indicating the charging time included in the usage record data in response to determining that charging has been performed in the determining;
• predicting a remaining capacity value of the battery at the time of return with respect to the usage schedule data acquired by the acquiring using a trained machine learning model that outputs a remaining capacity value of the battery at the time of return of the vehicle with the usage schedule data as an input;
• executing correction on the remaining capacity value of the battery at the time of return predicted by the predicting using the correction value calculated by the calculating;
• and
• outputting a remaining capacity value of the battery after correction by the executing.