Abstract: An automatic objects labeling method includes: capturing M consecutive image frames at one station of an assembly line. Performing an object detection step which includes selecting a detection image frame that displays an operation using a work piece against a target object from the M consecutive image frames; and calibrating the position range of the target object in the detection image frame; retracing from the detection image frame to select an Nth retraced image frame from the M consecutive image frames; obtaining a labeled image of the target object from the Nth retraced image frame according to the position range; comparing the labeled image with images of the M consecutive image frames to find at least one other labeled image similar to the target object; and storing both the labeled image and the at least one other labeled image as the same labeled data set.

Abstract: A multi-branch network architecture searching method includes: obtaining a training dataset with a plurality of input data; obtaining block design elements for blocks, wherein the blocks forms an architecture of a neural network, and the blocks are configured to perform a feature extraction on the input data to generate an output data; for each hyperparameter of the neural network, obtaining at least one hyperparameter setting value; inputting the training dataset, block design elements, and the at least one hyperparameter setting value into a hyperparameter optimization algorithm to generate a hyperparameter combination, wherein the hyperparameter combination includes one of the at least one hyperparameter setting corresponding to each hyperparameter; executing the neural network based on the hyperparameter combination and inputting a test dataset to evaluate a model performance of the neural network; and outputting the hyperparameter combination when the model performance reaches a threshold.

Abstract: A smart factory system for being set up between a smart factory and a backend-system provider is disclosed. The smart factory system includes: a factory installation, being installed in the smart factory; a plurality of sensors, being in connection with the factory installation; a smart machine box, being signally connected to each of the sensors; a computing and storing apparatus, being set up at premises of the backend-system provider; and a query-making apparatus. With the most cost-intensive functions like computing and storing set up and maintained by the backend-system provider, a smart factory can be easily started and run by locally setting up and maintaining sensors and smart machine boxes while remotely subscribing the functions maintained by the backend-system provider as services with payment. The system significantly reduces the costs for stating and running a smart factory, thereby encouraging transformation into or establishment of smart factories.

Abstract: A production schedule estimation method and a production schedule estimation system are provided. The production schedule estimation method includes the following steps. Current-day work-in-process data, machine group cycle time data of a machine group, and productivity data of the machine group are obtained. The current-day work-in-process data, the cycle time data of the machine group, and the productivity data of the machine group are inputted into a prediction model. Current-day cycle time data and a current-day move volume for each of multiple stations in the machine group are calculated through the prediction model. And, current-day move data is calculated according to the current-day cycle time data and the current-day move volume for each of the multiple stations in the machine group.

Abstract: An automatic objects labeling method includes: capturing M consecutive image frames at one station of an assembly line. Performing an object detection step which includes selecting a detection image frame that displays an operation using a work piece against a target object from the M consecutive image frames; and calibrating the position range of the target object in the detection image frame; retracing from the detection image frame to select an Nth retraced image frame from the M consecutive image frames; obtaining a labeled image of the target object from the Nth retraced image frame according to the position range; comparing the labeled image with images of the M consecutive image frames to find at least one other labeled image similar to the target object; and storing both the labeled image and the at least one other labeled image as the same labeled data set.

Abstract: A training device and a training method for a neural network model. The training method includes: obtaining a data set; completing, according to the data set, a plurality of artificial intelligence (AI) model trainings to generate a plurality of models corresponding to the plurality of AI model trainings respectively; selecting, according to a first constraint, a first model set from the plurality of models; and selecting, according to a second constraint, the neural network model from the first model set.

Abstract: An electronic device and a method for training a neural network model are provided. The method includes: obtaining a first neural network model and a first pseudo-labeled data; inputting the first pseudo-labeled data into the first neural network model to obtain a second pseudo-labeled data; determining whether a second pseudo-label corresponding to the second pseudo-labeled data matching a first pseudo-label corresponding to the first pseudo-labeled data; in response to the second pseudo-label matching the first pseudo-label, adding the second pseudo-labeled data to a pseudo-labeled dataset; and training the first neural network model according to the pseudo-labeled dataset.

Abstract: A production schedule estimation method and a production schedule estimation system are provided. The production schedule estimation method includes the following steps. Current-day work-in-process data, machine group cycle time data of a machine group, and productivity data of the machine group are obtained. The current-day work-in-process data, the cycle time data of the machine group, and the productivity data of the machine group are inputted into a prediction model. Current-day cycle time data and a current-day move volume for each of multiple stations in the machine group are calculated through the prediction model. And, current-day move data is calculated according to the current-day cycle time data and the current-day move volume for each of the multiple stations in the machine group.

Abstract: A training device and a training method for a neural network model are provided. The training method includes: obtaining a data set; completing, according to the data set, a plurality of artificial intelligence (AI) model trainings to generate a plurality of models corresponding to the plurality of AI model trainings respectively; selecting, according to a first constraint, a first model set from the plurality of models; and selecting, according to a second constraint, the neural network model from the first model set.

Abstract: A method for determining tool assignment preference applied to a semiconductor manufacturing system. At least one first tool and second tool and at least one first semiconductor process and second semiconductor process applied to the tools are provided. Demand moves provided by the first and second semiconductor processes are calculated. Assignment preferences of the first and second tools are determined using a statistical method. The statistical method is a two-step data feedback method, comprising the steps of, in the first step, calculating assignment preferences of tools without setting assignment preferences, and, in the second step, assigning assignment preferences to the first and second tools according to the calculation result in the first step, wherein the first tool is assigned to a first assignment preference with a lowest average utility rate, and the second tool is assigned to a second assignment preference.

Abstract: A method for determining tool assignment preference applied to a semiconductor manufacturing system. At least one first tool and second tool and at least one first semiconductor process and second semiconductor process applied to the tools are provided. Demand moves provided by the first and second semiconductor processes are calculated. Assignment preferences of the first and second tools are determined using a statistical method. The statistical method is a two-step data feedback method, comprising the steps of, in the first step, calculating assignment preferences of tools without setting assignment preferences, and, in the second step, assigning assignment preferences to the first and second tools according to the calculation result in the first step, wherein the first tool is assigned to a first assignment preference with a lowest average utility rate, and the second tool is assigned to a second assignment preference.