Abstract
Deep learning approaches are increasingly used to tackle forecasting tasks but require substantial training data. When samples are limited, synthetic data generation techniques can effectively augment datasets to improve model performance. Data augmentation is typically applied offline before training a model. However, when training with mini-batches, some batches may contain a disproportionate number of synthetic samples that do not align well with the original data characteristics. This work introduces an online data augmentation framework that generates synthetic samples during the training of neural networks. By creating synthetic samples for each batch alongside their original counterparts, we maintain a balanced representation between real and synthetic data throughout the training process. This approach fits naturally with the iterative nature of neural network training and eliminates the need to store large augmented datasets. We validated the proposed framework using 3797 time series from 6 benchmark datasets, three neural architectures, and seven synthetic data generation techniques. The experiments suggest that online data augmentation leads to better forecasting performance compared to offline data augmentation or no augmentation approaches. The framework and experiments are publicly available.

Abstract
Data augmentation is a crucial tool in time series forecasting, especially for deep learning architectures that require a large training sample size to generalize effectively. However, extensive datasets are not always available in real-world scenarios. Although many data augmentation methods exist, their limitations include the use of transformations that do not adequately preserve data properties. This paper introduces Grasynda, a novel graph-based approach for synthetic time series generation that: (1) converts univariate time series into a network structure using a graph representation, where each state is a node and each transition is represented as a directed edge; and (2) encodes their temporal dynamics in a transition probability matrix. We performed an extensive evaluation of Grasynda as a data augmentation method for time series forecasting. We use three neural network variations on six benchmark datasets. The results indicate that Grasynda consistently outperforms other time series data augmentation methods, including ones used in state-of-the-art time series foundation models. The method and all experiments are publicly available. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.

Abstract

Abstract

Abstract
Early fault detection remains a critical challenge in predictive maintenance (PdM), particularly within critical infrastructure, where undetected failures or delayed interventions can compromise safety and disrupt operations. Traditional anomaly detection methods are typically reactive, relying on real-time sensor data to identify deviations as they occur. This reactive nature often provides insufficient lead time for effective maintenance planning. To address this limitation, we propose a novel two-stage early detection framework that integrates time series forecasting with anomaly detection to anticipate equipment failures several hours in advance. In the first stage, future sensor signal values are predicted using forecasting models; in the second, conventional anomaly detection algorithms are applied directly to the forecasted data. By shifting from real-time to anticipatory detection, the framework aims to deliver actionable early warnings, enabling timely and preventive maintenance. We validate this approach through a case study focused on metro train systems, an environment where early fault detection is crucial for minimizing service disruptions, optimizing maintenance schedules, and ensuring passenger safety. The framework is evaluated across three forecast horizons (1, 3, and 6 hours ahead) using twelve state-of-the-art anomaly detection algorithms from diverse methodological families. Detection performance is assessed using five performance metrics. Results show that anomaly detection remains highly effective at short to medium horizons, with performance at 1-hour and 3-hour forecasts comparable to that of real-time data. Ensemble and deep learning models exhibit strong robustness to forecast uncertainty, maintaining consistent results with real-time data even at 6-hour forecasts. In contrast, distance- and density-based models suffer substantial degradation at longer horizons (6-hours), reflecting their sensitivity to distributional shifts in predicted signals. Overall, the proposed framework offers a practical and extensible solution for enhancing traditional PdM systems with proactive capabilities. By enabling early anomaly detection on forecasted data, it supports improved decision-making, operational resilience, and maintenance planning in industrial environments.

Abstract
One of the key aspects of Industry 4.0 is using intelligent systems to optimize manufacturing processes by improving productivity and reducing costs. These systems have greatly impacted in different areas, such as demand prediction and quality assessment. However, the prognostics and health management of industrial equipment is one of the areas with greater potential. This paper presents a comparative analysis of deep learning architectures applied to the prediction of the remaining useful life (RUL) on public real industrial datasets. The analysis includes some of the most commonly employed recurrent neural network variations and a novel approach based on a hybrid architecture using transformers. Moreover, we apply explainability techniques to provide comprehensive insights into the model's decision-making process. The contributions of the work are: (1) a novel transformer-based architecture for RUL prediction that outperforms traditional recurrent neural networks; (2) a detailed description of the design strategies used to construct the models on two under-explored datasets; (3) the use of explainability techniques to understand the feature importance and to explain the model's prediction and (4) making models built for reproducibility available to other researchers.