A behavior recognition device for recognizing behaviors of a semiconductor manufacturing apparatus includes a storage device and a control unit. The storage device is configured to store log data of the semiconductor manufacturing apparatus. The control unit is cooperatively connected to the storage device, and configured to build a transition state model based on the log data to analyze behaviors related to wafer transfer sequences and manufacturing operations of the semiconductor manufacturing apparatus.

The subject is to provide a technique of dealing with an internal state of a controller in an edge device by cooperation between an edge device and a computer such as a cloud server, and to generalize data communication for the cooperation and reduce an amount of data communication. Provided is a distributed system including an edge device and a diagnostic data computer. The edge device includes an in-edge controller including at least a processing unit, and a system element to be monitored by the in-edge controller. The processing unit diagnoses presence or absence of an abnormality in the processing unit due to an abnormality occurring in the system element, generates first diagnostic data indicating presence or absence of the abnormality of the processing unit, converts the first diagnostic data into second diagnostic data indicating a type of the abnormality of the processing unit, and transmits the second diagnostic data to the diagnostic data computer.

In a data obtaining step, a data obtaining unit may obtain certain data at a plurality of sampling times. The data may include measurement information regarding the amount of wear of the cutting tool and measurement information regarding a physical quantity of the cutting tool during cutting. In a clustering step, a clustering unit may cluster a plurality of pieces of the obtained data on the basis of the measurement information regarding the physical quantity included in each of the plurality of pieces of data. In a model creation step, a model creation unit may create, for each of clusters on the basis of the measurement information regarding the physical quantity and the measurement information regarding the amount of wear included in the data, a model for obtaining estimation information regarding the amount of wear from new measurement information regarding the physical quantity.

An information processing apparatus executes a simulation of a state of a process which is being performed in a semiconductor manufacturing apparatus, by using a simulation model of the semiconductor manufacturing apparatus. The information processing apparatus includes: a physical sensor data acquisition unit that acquires physical sensor data measured in the semiconductor manufacturing apparatus that is performing the process according to process parameters; a simulation execution unit that executes the simulation by the simulation model according to the process parameters, thereby outputting virtual sensor data; a simulation result determination unit that performs a pre-detection of a part of the semiconductor manufacturing apparatus that needs to be replaced, based on a difference between the physical sensor data and the virtual sensor data; and a part order unit that orders the part of the semiconductor manufacturing apparatus based on a result of the pre-detection.