https://www.mdu.se/

Artificial Intelligence (AI) is reshaping decision-making across industries, but in life-critical domains like Air Traffic Control (ATC), performance alone is not enough - AI must also be transparent, accountable, and human-centered. At the same time, ecological sustainability (e.g., fuel efficiency, emissions reduction, noise abatement) is becoming a key operational priority. Yet under high workload or time pressure, these goals are often deprioritized. This paper introduces a novel framework for bias-aware human-AI teaming in ATC, integrating Generative Explainable AI (genXAI), Predictive Telemetry (PT), multimodal machine learning, and mechanistic interpretability. Designed for Air Traffic Controllers (ATCOs), the system anticipates future traffic states, detects bias-prone conditions, and delivers adaptive, context-sensitive explanations. Through bias-aware interfaces, serious-game training, and neuroadaptive feedback loops, the framework aligns decision support with cognitive state and operational complexity. The framework targets three key objectives: (1) mitigating cognitive biases, (2) improving ground routing decisions, and (3) validating anticipatory AI models. By reframing sustainability as a human-AI collaboration challenge, this work advances a new class of trustworthy, bias-aware AI that enhances both safety and ecological performance in Air Traffic Management (ATM). 

Learning over time for machine learning (ML) models is emerging as a new field, often called continual learning or lifelong Machine learning (LML). Today, deep learning and neural networks are the prevalent approaches for LML models. However, they are often criticized for being “black box” methods, and catastrophic forgetting has remained a persistent challenge throughout their development. This paradigm represents a significant shift from traditional static learning models, enabling systems to adapt to new data continuously while retaining previously acquired knowledge. In this paper, an LML approach is proposed that combines a Random Forest (RF) with a Genetic Algorithm (GA) to transfer the knowledge from an existing RF model to a new learning model. Here, the GA is applied to the RF model so that the weights of this model get balanced more steadily. The approach is evaluated here for classification problems on three benchmark datasets. The initial results present knowledge retention in the new model, indicating the success of the model. These methods are regularization-based and effective in mitigating catastrophic forgetting. However, they rely on Fisher Information to estimate parameter importance or similar measures, which are computationally demanding and suitable for deep neural networks. While many NN-based lifelong learning approaches have been studied, RFs remain comparatively underexplored in this paradigm. Given their robustness, interpretability, and effectiveness in tabular datasets with limited samples, RFs present a compelling alternative. This motivates our work on developing an RF-based lifelong learning approach. 

Industrial environments generate large volumes of heterogeneous, unstructured PDF documents, such as material certificates, making reliable automated data extraction a methodological challenge. Traditional LLM-based pipelines often struggle with noisy OCR output, inconsistent layouts, and schema variability, leading to unstable outputs and reduced accuracy. This study addresses the problem of designing a robust, consistent extraction pipeline capable of operating across diverse document structures under both limited and scaled domain-specific supervision regimes. We propose a coordinated multi-agent framework integrating layout-aware OCR processing, expert-guided prompt engineering, systematic prompt refinement, controlled synthetic dataset expansion, and parameter-efficient adaptation via LoRA. The architecture includes specialised agents for raw text extraction, dataset preparation, prompt optimisation, LoRA configuration search, output processing, and structured evaluation, forming an iterative workflow that improves stability and reduces error propagation. Structural compliance and retrieval accuracy are evaluated separately using binomial modelling with Wilson confidence intervals. Zero-shot, few-shot, and LoRA fine-tuned LLMs are assessed across commercial and open-weight models. Results show that few-shot prompting enables rapid structural alignment under low-resource conditions, while fine-tuning effectiveness depends on supervision diversity. Merged supervision, defined as fine-tuning on combined original and synthetic certificates (FT-Merged), substantially outperforms limited-scale LoRA adaptation. Under FT-Merged, Mistral 7B achieves 97.05% retrieval accuracy with near-perfect structural compliance, approaching commercial reliability. GPT-4o and DeepSeek Chat achieve approximately 99% accuracy without task-specific fine-tuning. Cross-domain validation confirms structural generalisation, and runtime-cost analysis highlights trade-offs between accuracy, latency, financial cost, and deployment control. Overall, robust industrial extraction emerges from coordinated architectural orchestration rather than model choice alone. © 2026

Explainability is crucial for establishing user trust in Artificial Intelligence (AI), particularly within safety-critical domains such as Air Traffic Management (ATM) and Air Traffic Control (ATC). This study empirically investigates the effects of Explainable AI (XAI), specifically HeatMap-based visual explanations, on cognitive workload, user acceptance, and intention to use AI-driven decision-support systems among Air Traffic Control Officers (ATCOs). Despite significant theoretical advancements in the broader XAI domain, empirical evidence addressing the specific impact of visual explanations on human-AI interactions in safety-critical environments like ATC remains limited. To address these critical gaps, an experimental comparison was conducted between explainable (HeatMap) and non-explainable (BlackBox) AI conditions, involving two user groups: expert and student ATCOs. Both objective neurophysiological measures (Electroencephalography) and subjective questionnaires were employed to capture comprehensive user responses. Key findings revealed that the presence of visual explanations significantly reduced cognitive workload and enhanced users' willingness to adopt the AI system, regardless of participants' level of expertise. However, explicit perceptions of AI's impact on work performance were predominantly influenced by expertise, with less experienced controllers reporting a greater perceived impact than their expert counterparts. By combining objective neurometrics with subjective user assessments, this research advances methodological rigor in evaluating human-AI interactions and highlights the importance of tailored, user-centric explanations. These findings directly contribute to practical guidelines for designing cognitively compatible and trustworthy AI tools in ATC, providing nuanced insights for targeted training and deployment strategies based on user expertise. 

Predictive maintenance is critical for minimizing downtime and operational costs in mining industries, where slurry pumps operate under abrasive and highly variable conditions. Traditional machine learning models, while accurate, often lack interpretability, limiting their adoption in safety-critical environments. This paper presents an Adaptive Neuro-Fuzzy Inference System (ANFIS) based approach for anomaly detection in slurry pumps using vibration-based features such as displacement, velocity, crest factor, kurtosis, skewness, peak, and peak-to-peak values. ANFIS combines the transparency of fuzzy logic with the learning capability of neural networks, enabling interpretable IF-THEN rules and adaptive tuning of membership functions. A dataset of 33,104 vibration samples was analyzed under four balancing strategies to address class imbalance. Models were evaluated using accuracy, precision, recall, F1-score, RMSE, and clustering quality metrics (Dunn Index and Silhouette Score). The results demonstrate that ANFIS achieves high accuracy (>99%) and strong interpretability, outperforming traditional black-box models. The proposed approach enhances trustworthiness and adaptability in predictive maintenance systems, offering an explainable solution for Industry 4.0 applications. Future work will explore hybrid models, real-time IoT integration, and edge deployment for dynamic operational environments.

High-voltage direct current (HVDC) systems rely on extensive and complex cooling networks that incorporate both dry coolers and chillers. Designing these systems is very challenging because many parameters interact in nonlinear ways, and simulation-based optimization requires significant computational resources. This paper presents a machine learning framework incorporating physics-based constraints for the efficient optimization of parametric cooling systems. The framework includes three main stages: data exploration, modeling, and optimization. In the first stage, simulation data are analyzed to extract key parameters such as Cooling Efficiency, Power Consumption, Cooling Power, Chiller Capacity, Water Inlet Temperature, etc. The modeling stage uses machine learning models to predict a composite performance score that combines cooling effectiveness, energy efficiency, and CO2 savings. Finally, a physics-constrained optimization algorithm (L-BFGS-B: Limited-memory Broyden–Fletcher–Goldfarb–Shanno with Bounds) finds the best parameter combinations while ensuring physical consistency in energy balance, temperature limits, and COP values. The results show that the proposed AI model can accurately predict system performance and identify optimal configurations. For 20 system configurations, the model achieved an R2 of 0.93, showing strong predictive capability. Key influential factors include Fan Power, Cooling Efficiency, COP (Chiller) and Power Consumption. This approach has the potential to provide a accurate and reliable alternative to full-scale simulations.

Trustworthy decision support systems utilizing a multimodal approach (MMA) integrate diverse data modalities to enhance robustness, transparency, and fairness in artificial intelligence (AI) applications. In this study, we present an MMA for decision support in the Air Traffic Management (ATM) domain, particularly within Remote Digital Towers (RDTs). RDTs replace traditional control towers with AI-driven digital solutions, enhancing operational efficiency. Our approach addresses key multimodal challenges—translation, alignment, and co-learning—by implementing (a) an open-vocabulary-based object detection model for video processing and (b) an audio-to-text transcription and semantic word identification model. The YOLO-World deep-learning model is employed for object detection, while audio data analysis takes advantage of a benchmark data set, semantic identification techniques, and explainability. Additionally, the system integrates robust machine learning techniques, including data augmentation and perturbation, to maintain consistent performance across varied operational conditions. This proof-of-concept demonstrates the potential of multimodal AI systems to enhance decision support and improve safety in ATM environments.

This study uses Generative Artificial Intelligence (gAI) to advance industrial digitization. Although the use of gAI looks promising for industrial digitization, there are significant gaps in current Explainable Artificial Intelligence (XAI) methods, which limit their applicability to such applications. By developing a theoretical framework, the aim is to provide explanations for gAI to improve decision-making processes with actionable insights and explanations for their intended outcomes. The proposed work has an impact on facilitating inspection, monitoring, optimization, and maintenance of industrial equipment and machinery. The theoretical framework proposed in this paper will address this challenge by following a three-step approach: 1) learning prior and posterior from data, 2) feature attribution and counterfactual explanation-based methods, and 3) integrated XAI. While the current study is theoretical, future work will focus on applying the approach to real-world industrial scenarios.

This work presents an explainable AI framework for fault predictions of Embedded Test Systems, combining classification and trend prediction. Random Forest (RF) and Gradient Boosting (GB) are used as baselines for classification, and the pipeline is extended with sliding-window sequence modelling using LSTM. With the data set, RF achieved 99.84% accuracy and 1.00 ROC-AUC, while Gradient Boosting achieved 98.59% accuracy and 0.96 ROC-AUC. The LSTM forecasts nexttimestep measurement and supports control-chart monitoring, yielding low errors for passed instances (MAE ≈0.007; RMSE ≈0.008) and higher errors for failed ones (MAE ≈1.158; RMSE ≈1.454), effectively flagging unstable behaviour. To enhance interpretability, SHAP and LIME explanations are computed and deployed in a Django-based web application for uploading data, prediction, and visualisation. Additionally, a lightweight Large Language Model (LLM) generates natural language rationales, helping engineers understand which features and time segments drive each decision. Results confirm that tree-based models provide robust baselines, while sequence-aware modelling and explainable AI add practical value for monitoring signals in production.

Annotating vibration data from heavy-duty pumps in the mining industry is highly challenging because it demands domain knowledge, a complex inspection setup, and, in many cases, remains infeasible. A self-supervised data annotation (SSDA) framework is therefore proposed and evaluated on historical data of slurry-pump vibration signals. The framework began with the collection of heterogeneous information, followed by information fusion using an autoencoder. This was then followed by a datafication step for preprocessing and achieving a better representation of features through a feature embedding technique. As a result, redundant information was pushed into an eight-dimensional latent space, achieving a reconstruction loss of 0.0023. Furthermore, Initial data annotation was obtained by combining the Isolation Forest and Kneedle algorithms to locate a data-driven knee or threshold, and it was found to be 0.58 for predicting labels. Partial samples were labeled and considered accurate. Lastly, an attention-based fuzzy neural network (AFNN) is trained on those labels where membership functions convert each latent feature into graded truth values. At the same time, an attention layer highlights the most relevant rules. An iterative self-training loop was implemented to refine the training set and obtain labeled data with higher model confidence. Here, we also tested six baseline models and found AFNN quite impressive. After seven iterations 2780 of 2872 samples were labeled and the remaining 92 are considered uncertain, still need some review from an expert, and the AFNN model confidence was (96.8%). Statistical analysis confirmed that the model predictions were significantly associated with true labels (p<0.05) and not driven by chance.