公开(公告)号：US20220180149A1

公开(公告)日：2022-06-09

申请号：US17472483

申请日：2021-09-10

申请人： KABUSHIKI KAISHA TOSHIBA

发明人： Yasuo NAMIOKA ,  Atsushi WADA ,  Takanori YOSHII

IPC分类号： G06N3/04

摘要： According to one embodiment, a training device trains a first model. The first model estimates a period of a task from time-series data of an operation of a human. The device acquires first time-series data to which a label of the task is assigned. The device extracts a pattern from a period indicated by the label in the first time-series data. The pattern is used as a feature. The device generates timing data of an appearance timing of the pattern in the first time-series data. The device trains the first model by using the label, the first time-series data, and the timing data.

公开(公告)号：US20230085797A1

公开(公告)日：2023-03-23

申请号：US17653193

申请日：2022-03-02

申请人： KABUSHIKI KAISHA TOSHIBA

发明人： Takanori YOSHII ,  Hiroaki NAKAMURA ,  Takehiro KATO ,  Makoto TSUJI ,  Yasuo NAMIOKA

IPC分类号： G05B19/4093

摘要： According to one embodiment, an analysis device performs an analysis related to a plurality of tasks of a manufacturing process. The analysis device receives an image when each of the plurality of tasks is performed. The analysis device receives the images from an imaging device acquiring the images. The analysis device receives a detection signal from a tool used in at least one of the plurality of tasks. The detection signal is detected by the tool. The analysis device refers to end determination data for determining an end of each of the plurality of tasks. The analysis device determines the end of each of the plurality of tasks based on the images, the detection signal, and the end determination data.