https://www.mdu.se/

Reasoning is a vital process in machine learning (ML), involving making inferences and drawing conclusions based on data. This capability is important for developing intelligent systems which can understand and predict complex patterns. The study investigates reasoning through two distinct methodologies: multimodal reasoning and transfer learning based reasoning. In the first approach, multimodal reasoning is used with a semi-supervised method to label unlabelled datasets. In the second approach, transfer learning has been used to transfer knowledge of data from one model to another. Both approaches are demonstrated using unlabelled vehicular telemetry data. During processing, three sets of telemetry data are used to extract features separately through the autoencoder. These features are then clustered and aligned to create labelled and unlabelled datasets. The eXtreme Gradient Boosting (XGBoost) algorithm achieved over 98% test accuracy when applied to the labelled datasets and was then used to predict labels for the unlabelled datasets, which were later added to the labelled dataset to form three datasets for further processing. In transfer learning, a transformer model specifically designed to handle continuous features is developed. Labelled datasets are applied to the transformer model, one after the other, resulting in three final models, with each model achieving over 80% accuracy. The model’s prediction confidence is also validated using conformal learning, where the final models achieved over 80% accuracy. The transformer model is also separately trained on the datasets and compared with traditional ML models, outperforming the others by achieving an accuracy of 98%. By building on the groundwork laid by this study, future research can push the boundaries of what is possible with reasoning approaches, opening up new paths for scientific exploration and practical applications in different fields.

Ontology development plays a vital role as it provides a structured way to represent and organize knowledge. It has the potential to connect and integrate data from different sources, enabling a new class of AI-based services and systems such as decision support systems and recommender systems. However, in large manufacturing industries, the development of such ontology can be challenging. This paper presents a use case of an application ontology development based on machine breakdown work orders coming from a Computerized Maintenance Management System (CMMS). Here, the ontology is developed using a Knowledge Meta Process: Methodology for Ontology-based Knowledge Management. This ontology development methodology involves steps such as feasibility study, requirement specification, identifying relevant concepts and relationships, selecting appropriate ontology languages and tools, and evaluating the resulting ontology. Additionally, this ontology is developed using an iterative process and in close collaboration with domain experts, which can help to ensure that the resulting ontology is accurate, complete, and useful for the intended application. The developed ontology can be shared and reused across different AI systems within the organization, facilitating interoperability and collaboration between them. Overall, having a well-defined ontology is critical for enabling AI systems to effectively process and understand information.

The fourth industrial revolution, colloquially referred to as “industry 4.0”, has garnered substantial global attention in recent years. There, Artificial intelligence (AI) driven industrial intelligence has been increasingly deployed in predictive maintenance (PdM), emerging as a vital enabler of smart manufacturing and industry 4.0. Since in recent years the number of articles focusing on Artificial Intelligence (AI) in PdM is high a review on the available literature reviews in this domain would be useful for the future researchers who would like to advance the research in this area and also for the persons who would like to apply PdM in their application domains. Therefore, this study identifies the AI revolution in PdM and focuses on the next stages available in the literature reviews in this area by quality assessment of secondary study. A well-known structured review approach (Systematic Literature Review, or SLR) was employed to perform this tertiary study. In addition, the Scale for the Assessment of Narrative Review Articles (SANRA) approach for evaluating the quality of review papers has been employed to support a few of the research questions. Here, This tertiary study scrutinizes four crucial aspects of secondary articles: (1) their specific research domains, (2) the annual trends in the quantity, variety, and quality (3) a footsteps of top researchers, and (4) the research constraints that review articles face during the time frame of 2015 to 2022. The results show that the majority of the application areas are applied to the manufacturing industry. It also leads to the identification of the revolution of AI in PdM as well. Our final findings indicate that Dr. Cheng et al.’s (2022) review has emerged as the predominant source of information in this field. As newcomers or industrial practitioners, we can benefit greatly from following his insights. The final outcome is that there is a lack of progress in SLR formulation and in adding explainable or interpretive AI methodologies in secondary studies.

In the contemporary landscape of decision support systems, machine learning (ML) algorithms assume a pivotal role in diverse domains, including job screening and loan approvals. Despite their extensive utilization, a persistent challenge arises in the form of biased outcomes, notably influenced by sensitive attributes such as gender and ethnicity. While current research heavily leans on these attributes for fairness, the scarcity of data due to privacy and legal constraints poses a substantial hurdle. Furthermore, imbalances in real-world datasets necessitate the use of class balancing techniques, but conflicting findings on their impact on bias mitigation and overall model performance complicate the pursuit of fairness. This paper conducts a comprehensive investigation, addressing the unique challenge of constructing fair models without explicit reliance on sensitive attributes. It specifically examines the effectiveness of Synthetic Minority Oversampling TEchnique (SMOTE)-driven oversampling methods. The study's findings reveal a significant enhancement in classification performance through SMOTE-driven techniques. These insights advocate for the thoughtful integration of SMOTE-driven oversampling techniques to achieve a balance between model fairness and accuracy. The results provide valuable guidance to researchers and practitioners in the field, contributing to the ongoing dialogue on fairness in machine learning models. 

Recent strides in integrating artificial intelligence (AI) with production systems align with the trend towards highly automated manufacturing, demanding smarter machinery. This dovetails with the overarching vision of Industry 4.0, moving beyond conventional models towards employing AI for real-time modeling of production processes, enabling adaptable and learning-enabled models. This study focuses on leveraging cutting-edge deep learning techniques to monitor and classify tool wear using authentic image data from machining processes. Various deep learning algorithms, including CNN, AlexNet, EfficientNetB0, MobileNetV2, CoAtNet-0, and ResNet18, are explored for monitoring and measuring wear through images of machining chips. The collected images of machining chips are categorized as ‘Accepted’, ‘Unaccepted’, and ‘Optimal’. Due to imbalanced datasets, the study investigates two distinct strategies: upsampling and downsampling. The study also aimes to enhance sensitivity for a specific minority class to meet industrial requirements. The study showed that upsampling enhanced accuracy and almost fulfilled the stated requirements, whereas downsampling did not achieve the desired outcomes. The study evaluates and compares the effectiveness of recently introduced deep learning algorithms with other CNN-based architectures in classifying tool wear states in real-world scenarios. It sheds light on the challenges faced by the machining industry, particularly the prevalent issue of class imbalance in real-world machining data. The observed results indicate that ResNet18 and AlexNet outperform other algorithms, achieving a weighted average accuracy of 96% for both multiclass and binary classification problems when considering upsampled datasets. Consequently, the study concludes that both ResNet18 and AlexNet demonstrate adaptability to class imbalances, generalization to real-world machining scenarios, and competitive accuracy.

Companies must enhance total maintenance effectiveness to stay competitive, focusing on both digitalization and basic maintenance procedures. Digitalization offers technologies for data-driven decision-making, but many maintenance decisions still lack a factual basis. Prioritizing efficiency and effectiveness require analyzing equipment history, facilitated by using Computerized Maintenance Management Systems (CMMS). However, CMMS data often contains unstructured free-text, leading to manual analysis, which is resource-intensive and reactive, focusing on short time periods and specific equipment. Two approaches are available to solve the issue: minimizing free-text entries or using advanced methods for processing them. Free-text allows detailed descriptions but may lack completeness, while structured reporting aids automated analysis but may limit fault description richness. As knowledge and experience are vital assets for companies this research uses a hybrid approach by combining Natural Language Processing with domain specific ontology and Large Language Models to extract information from free-text entries, enabling the possibility of real-time analysis e.g., identifying recurring failure and knowledge sharing across global sites.

Artificial Intelligence (AI) has recently made significant advancements and is now pervasive across various application domains. This holds true for Air Transportation as well, where AI is increasingly involved in decision-making processes. While these algorithms are designed to assist users in their daily tasks, they still face challenges related to acceptance and trustworthiness. Users often harbor doubts about the decisions proposed by AI, and in some cases, they may even oppose them. This is primarily because AI-generated decisions are often opaque, non-intuitive, and incompatible with human reasoning. Moreover, when AI is deployed in safety-critical contexts like Air Traffic Management (ATM), the individual decisions generated by AI models must be highly reliable for human operators. Understanding the behavior of the model and providing explanations for its results are essential requirements in every life-critical domain. In this scope, this project aimed to enhance transparency and explainability in AI algorithms within the Air Traffic Management domain. This article presents the results of the project’s validation conducted for a Conflict Detection and Resolution task involving 21 air traffic controllers (10 experts and 11 students) in En-Route position (i.e. hight altitude flight management). Through a controlled study incorporating three levels of explanation, we offer initial insights into the impact of providing additional explanations alongside a conflict resolution algorithm to improve decision-making. At a high level, our findings indicate that providing explanations is not always necessary, and our project sheds light on potential research directions for education and training purposes.

Tree-ensemble models, such as Extreme Gradient Boosting (XGBoost), are renowned Machine Learning models which have higher prediction accuracy compared to traditional tree-based models. This higher accuracy, however, comes at the cost of reduced interpretability. Also, the decision path or prediction rule of XGBoost is not explicit like the tree-based models. This paper proposes the iXGB--interpretable XGBoost, an approach to improve the interpretability of XGBoost. iXGB approximates a set of rules from the internal structure of XGBoost and the characteristics of the data. In addition, iXGB generates a set of counterfactuals from the neighbourhood of the test instances to support the understanding of the end-users on their operational relevance. The performance of iXGB in generating rule sets is evaluated with experiments on real and benchmark datasets which demonstrated reasonable interpretability. The evaluation result also supports that the interpretability of XGBoost can be improved without using surrogate methods.

The development and deployment of AI systems necessitate a steadfast commitment to reliability, safety, security, ethics, and social responsibility. This paper introduces key research issues and challenges for trustworthy AI based on our experience working on several ongoing research projects at Mälardalen University (MDU), Sweden, which considers practical, real-world scenarios from the mobility, transportation, and healthcare domains. Our observations have highlighted several critical technical components that underpin trustworthy AI. These components include fairness, safety, transparency, explainability, accountability, rigorous testing, verification, and a human-centric approach to AI. Notably, these elements align closely with the current state-of-the-art practices in the field.

Multimodal machine learning is a critical aspect in the development and advancement of AI systems. However, it encounters significant challenges while working with multimodal data, where one of the major issues is dealing with unlabelled multimodal data, which can hinder effective analysis. To address the challenge, this paper proposes a multimodal reasoning approach adopting second-order learning, incorporating grounding alignment and semi-supervised learning methods. The proposed approach illustrates using unlabelled vehicular telemetry data. During the process, features were extracted from unlabelled telemetry data using an autoencoder and then clustered and aligned with true labels of neurophysiological data to create labelled and unlabelled datasets. In the semi-supervised approach, the Random Forest (RF) and eXtreme Gradient Boosting (XGBoost) algorithms are applied to the labelled dataset, achieving a test accuracy of over 97%. These algorithms are then used to predict labels for the unlabelled dataset, which is later added to the labelled dataset to retrain the model. With the additional prior labelled data, both algorithms achieved a 99% test accuracy. Confidence in predictions for unlabelled data was validated using counting samples based on the prediction score and Bayesian probability. RF and XGBoost scored 91.26% and 97.87% in counting samples and 98.67% and 99.77% in Bayesian probability, respectively.