Abstract: There is provided a model selection apparatus including: an evaluation model storage unit configured to store each of a plurality of evaluation models capable of outputting an index for evaluating a state of a facility that is configured to manufacture a product from a raw material, in association with the raw material; a property data acquisition unit configured to acquire property data indicating a property of the raw material which is used in the facility; a model selection unit configured to select a target model for evaluating the state of the facility based on the property data, from among the plurality of evaluation models, when a target raw material in the facility is used; and a target model output unit configured to output the target model.

Abstract: An evaluation model generating apparatus, including: a property data obtaining unit which obtains property data representing a property of a raw material to be used in equipment that manufactures a product from the raw material; a determining unit which determines, based on the property data, whether to generate an evaluation model that outputs an indicator obtained by evaluating a state of the equipment; and an evaluation model generating unit which generates, according to a result of the determining, the evaluation model by machine learning, is provided. In the evaluation model generating apparatus, the determining unit may determine, if it is judged that a target raw material is not similar in property to a raw material that has been used in the past based on the property data, to generate the evaluation model.

Abstract: An estimation apparatus is provided, including an abnormality index acquisition unit for acquiring, as an abnormality index, an evaluation index output by an evaluation model according to state data being input when an abnormality occurs in a facility, among the state data indicating a state of the facility when an operation model is used to control a control target, the operation model being generated by reinforcement learning where an output of the evaluation model trained by machine learning to output the evaluation index in accordance with the state of the facility is set as at least a part of a reward, and outputting an action in accordance with the state of the facility; an estimation unit for estimating which of the evaluation model or the operation model is a main cause of the abnormality; and an output unit for executing an output in accordance with a result of the estimate.

Abstract: Provided is a model selection apparatus including a candidate model storage unit configured to store a plurality of candidate models each of which is generated by reinforcement learning using, as at least a part of a reward, an output of an evaluation model configured to output an indicator obtained by evaluating a state of equipment and is capable of output an action, a state data acquisition unit configured to acquire a plurality of pieces of state data representing the state of the equipment in a case where each of manipulated variables based on outputs of the plurality of candidate models is applied to a controlled object in the equipment, an indicator acquisition unit configured to acquire a plurality of indicators, a model selection unit configured to select an object model for controlling the controlled object, and an object model output unit configured to output the object model.

Abstract: Provided is a model verification apparatus including a model storage unit configured to store a machine learning model that has been subjected to machine learning, by using a learning sample including state data indicating a state of equipment in which a controlled object is provided and action data indicating an action for deciding a manipulated variable to be applied to the controlled object, so as to output an action according to the state of the equipment, a learning range identification unit configured to identify a learning range indicating a range of the learning sample, a verification data acquisition unit configured to acquire verification data indicating a plurality of states of the equipment outside the learning range, and a verification result output unit configured to output a verification result indicating a plurality of actions to be output by the machine learning model in response to an input of the verification data.

Abstract: Provided is an apparatus comprising an acquisition portion for acquiring measurement data measured for a state of an object; a storage portion for storing a model that outputs deterioration strength index value indicating strength of causing the object to deteriorate, and a deterioration parameter indicating a correspondence between the deterioration strength index value and a progression rate of deterioration of the object, in response to an input of the measurement data; and an estimation portion for estimating a progression rate of deterioration of the object by using the deterioration strength index value outputted in response to inputting the measurement data into the model, and the deterioration parameter.