The dynamicity of real-world systems poses a significant challenge to deployed predictive machine learning (ML) models. Changes in the system on which the ML model has been trained may lead to performance degradation during the system’s life cycle. Recent advances that study non-stationary environments have mainly focused on identifying and addressing such changes caused by a phenomenon called concept drift. Different terms have been used in the literature to refer to the same type of concept drift and the same term for various types. This lack of unified terminology is set out to create confusion on distinguishing between different concept drift variants. In this paper, we start by grouping concept drift types by their mathematical definitions and survey the different terms used in the literature to build a consolidated taxonomy of the field. We also review and classify performance-based concept drift detection methods proposed in the last decade. These methods utilize the predictive model’s performance degradation to signal substantial changes in the systems. The classification is outlined in a hierarchical diagram to provide an orderly navigation between the methods. We present a comprehensive analysis of the main attributes and strategies for tracking and evaluating the model’s performance in the predictive system. The paper concludes by discussing open research challenges and possible research directions.

Most machine learning methods assume that the input data distribution is the same in the training and testing phases.However, in practice, this stationarity is usually not met and the distribution of inputs differs, leading to unexpectedperformance of the learned model in deployment. The issue in which the training and test data inputs follow differentprobability distributions while the input–output relationship remains unchanged is referred to as covariate shift. In thispaper, the performance of conventional machine learning models was experimentally evaluated in the presence of covariateshift. Furthermore, a region-based evaluation was performed by decomposing the domain of probability density function ofthe input data to assess the classifier’s performance per domain region. Distributional changes were simulated in a twodimensional classification problem. Subsequently, a higher four-dimensional experiments were conducted. Based on theexperimental analysis, the Random Forests algorithm is the most robust classifier in the two-dimensional case, showing thelowest degradation rate for accuracy and F1-score metrics, with a range between 0.1% and 2.08%. Moreover, the resultsreveal that in higher-dimensional experiments, the performance of the models is predominantly influenced by the complexity of the classification function, leading to degradation rates exceeding 25% in most cases. It is also concluded that themodels exhibit high bias toward the region with high density in the input space domain of the training samples.

Load forecasting is a crucial topic in energy management systems (EMS) due to its vital role in optimizing energy scheduling and enabling more flexible and intelligent power grid systems. As a result, these systems allow power utility companies to respond promptly to demands in the electricity market. Deep learning (DL) models have been commonly employed in load forecasting problems supported by adaptation mechanisms to cope with the changing pattern of consumption by customers, known as concept drift. A drift magnitude threshold should be defined to design change detection methods to identify drifts. While the drift magnitude in load forecasting problems can vary significantly over time, existing literature often assumes a fixed drift magnitude threshold, which should be dynamically adjusted rather than fixed during system evolution. To address this gap, in this paper, we propose a dynamic drift-adaptive Long Short-Term Memory (DA-LSTM) framework that can improve the performance of load forecasting models without requiring a drift threshold setting. We integrate several strategies into the framework based on active and passive adaptation approaches. To evaluate DA-LSTM in real-life settings, we thoroughly analyze the proposed framework and deploy it in a real-world problem through a cloud-based environment. Efficiency is evaluated in terms of the prediction performance of each approach and computational cost. The experiments show performance improvements on multiple evaluation metrics achieved by our framework compared to baseline methods from the literature. Finally, we present a trade-off analysis between prediction performance and computational costs.

Most industrial processes in real-world manufacturing applications are characterized by the scalability property, which requires an automated strategy to self-adapt machine learning (ML) software systems to the new conditions. In this paper, we investigate an Electroslag Remelting (ESR) use case process from the Uddeholms AB steel company. The use case involves predicting the minimum pressure value for a vacuum pumping event. Taking into account the long time required to collect new records and efficiently integrate the new machines with the built ML software system. Additionally, to accommodate the changes and satisfy the non-functional requirement of the software system, namely adaptability, we propose an automated and adaptive approach based on a drift handling technique called importance weighting. The aim is to address the problem of adding a new furnace to production and enable the adaptability attribute of the ML software. The overall results demonstrate the improvements in ML software performance achieved by implementing the proposed approach over the classical non-adaptive approach.