Monitoring the status of the facilities and detecting any faults are considered an important technology in a smart factory. Although the faults of machine can be analyzed in real time using collected data, it requires a large amount of computing resources to handle the massive data. A cloud server can be used to analyze the collected data, but it is more efficient to adopt the edge computing concept that employs edge devices located close to the facilities. Edge devices can improve data processing and analysis speed and reduce network costs. In this paper, an edge device capable of collecting, processing, storing and analyzing data is constructed by using a single-board computer and a sensor. And, a fault detection model for machine is developed based on the long short-term memory (LSTM) recurrent neural networks. The proposed system called LiReD was implemented for an industrial robot manipulator and the LSTM-based fault detection model showed the best performance among six fault detection models.

Recently, convolutional neural network (CNN)-based classification models have shown good performance for motor imagery (MI) brain-computer interfaces (BCI) using electroencephalogram (EEG) in end-to-end learning. Although a few explainable artificial intelligence (XAI) techniques have been developed, it is still challenging to interpret the CNN models for EEG-based BCI classification effectively. In this research, we propose 3D-EEGNet as a 3D CNN model to improve both the explainability and performance of MI EEG classification. The proposed approach exhibited better performances on two MI EEG datasets than the existing EEGNet, which uses a 2D input shape. The MI classification accuracies are improved around 1.8% and 6.1% point in average on the datasets, respectively. The permutation-based XAI method is first applied for the reliable explanation of the 3D-EEGNet. Next, to find a faster XAI method for spatio-temporal explanation, we design a novel technique based on the normalized discounted cumulative gain (NDCG) for selecting the best among a few saliency-based methods due to their higher time complexity than the permutation-based method. Among the saliency-based methods, DeepLIFT was selected because the NDCG scores indicated its results are the most similar to the permutation-based results. Finally, the fast spatio-temporal explanation using DeepLIFT provides deeper understanding for the classification results of the 3D-EEGNet and the important properties in the MI EEG experiments.

Today, real-time fault detection and predictive maintenance based on sensor data are actively introduced in various areas such as manufacturing, aircraft, and power system monitoring. Many faults in motors or rotating machinery like industrial robots, aircraft engines, and wind turbines can be diagnosed by analyzing signal data such as vibration and noise. In this study, to detect failures based on vibration data, preprocessing was performed using signal processing techniques such as the Hamming window and the cepstrum transform. After that, 10 statistical condition indicators were extracted to train the machine learning models. Specifically, two types of Mahalanobis distance (MD)-based one-class classification methods, the MD classifier and the Mahalanobis–Taguchi system, were evaluated in detecting the faults of rotating machinery. Their performance for fault detection on rotating machinery was evaluated with different imbalanced ratios of data by comparing with binary classification models, which included classical versions and imbalanced classification versions of support vector machine and random forest algorithms. The experimental results showed the MD-based classifiers became more effective than binary classifiers in cases in which there were much fewer defect data than normal data, which is often common in the real-world industrial field.

Worker safety in modern industrial environments is becoming increasingly critical owing to the complexities introduced by human–machine collaborations, making real-time activity recognition essential. Conventional unimodal human activity recognition (HAR), which relies solely on inertial sensors or cameras, often exhibits degraded performance in dynamic settings owing to occlusion, sensor faults, and poor scalability in multi-worker scenarios. To address these limitations, we propose a real-time multimodal HAR system as the core component of a human digital twin (HDT) for safety management. The system integrates inertial measurement unit (IMU) data from workers’ smartphones with vision data from existing closed-circuit television (CCTV) infrastructure, eliminating the need for additional wearables. Worker identity is maintained by linking smartphone identifiers with face recognition results obtained from video streams, enabling consistent tracking even when workers temporarily leave the camera’s field of view. An adaptive score fusion strategy dynamically balances modalities according to confidence, which improves the robustness of the system to missing or noisy data. This approach advances HAR beyond controlled laboratory settings and provides a practical and computationally efficient foundation for HDT-based safety management in industrial environments.