Handling Imbalanced Datasets for Robust Deep Neural Network-Based Fault Detection in Manufacturing Systems

Abstract

Over the recent years, Industry 4.0 (I4.0) technologies such as the Industrial Internet of Things (IIoT), Artificial Intelligence (AI), and the presence of Industrial Big Data (IBD) have helped achieve intelligent Fault Detection (FD) in manufacturing. Notably, data-driven approaches in FD apply Deep Learning (DL) techniques to help generate insights required for monitoring complex manufacturing processes. However, due to the ratio of instances where actual faults occur, FD datasets tend to be imbalanced, leading to training challenges that result in inefficient DL-based FD models. In this paper, we propose Dual Logits Weights Perturbation (DLWP) loss, a method featuring weight vectors for improved dataset generalization in FD systems. The weight vectors act as hyperparameters adjusted on a case-by-case basis to regulate focus accorded to individual minority classes during training. In particular, our proposed method is suitable for imbalanced datasets from safety-related FD tasks as it generates DL models that minimize false negatives. Subsequently, we integrate human experts into the workflow as a strategy to help safeguard the system. A subset of the results, model predictions with uncertainties exceeding a preset threshold, are considered a preliminary output subject to cross-checking by human experts. We demonstrate that DLWP achieves improved Recall, AUC, F1 scores.

Automated Cow Body Condition Scoring Using Multiple 3D Cameras and Convolutional Neural Networks

“…Automated body condition scoring research knows this problem all too well with the majority of the training samples sitting in the middle classes and a small number of samples in the edge classes, as can be seen in Figure 3. There are numerous techniques available for dealing with imbalanced classes such as oversampling under-represented classes, undersampling over-represented classes, cost-sensitive learning, and even transfer learning [27]. The technique chosen for this study was cost-sensitive learning and the approach is explained in detail in Section 3.5.3.…”

“…As mentioned before, there are numerous techniques available for dealing with imbalanced classes such as oversampling under-represented classes, undersampling overrepresented classes, cost-sensitive learning, and even transfer learning [27]. Cost-sensitive learning was the chosen technique for this study.…”

Enhanced Dual Convolutional Neural Network Model Using Explainable Artificial Intelligence of Fault Prioritization for Industrial 4.0

“…The Quality 4.0 branch of I4.0 is present. By utilizing cutting-edge computational tools and new approaches, this field seeks to improve product quality [3,4]. There are many social Sensors 2023, 23, 7011 2 of 23 constructivisms and patterns, but the main obstacle is grasping Grade 4.0 in social studies.…”