Although the effectiveness of machine learning (ML) for machine diagnosis has been widely established, the interpretation of the diagnosis outcomes is still an open issue. Machine learning models behave as black boxes; therefore, the contribution given by each of the selected features to the diagnosis is not transparent to the user. This work is aimed at investigating the capabilities of the SHapley Additive exPlanation (SHAP) to identify the most important features for fault detection and classification in condition monitoring programs for rotating machinery. The authors analyse the case of medium-sized bearings of industrial interest. Namely, vibration data were collected for different health states from the test rig for industrial bearings available at the Mechanical Engineering Laboratory of Politecnico di Torino. The Support Vector Machine (SVM) and k-Nearest Neighbour (kNN) diagnosis models are explained by means of the SHAP. Accuracies higher than 98.5% are achieved for both the models using the SHAP as a criterion for feature selection. It is found that the skewness and the shape factor of the vibration signal have the greatest impact on the models’ outcomes.

Today’s deep learning strategies require ever-increasing computational efforts and demand for very large amounts of labelled data. Providing such expensive resources for machine diagnosis is highly challenging. Transfer learning recently emerged as a valuable approach to address these issues. Thus, the knowledge learned by deep architectures in different scenarios can be reused for the purpose of machine diagnosis, minimizing data collecting efforts. Existing research provides evidence that networks pre-trained for image recognition can classify machine vibrations in the time-frequency domain by means of transfer learning. So far, however, there has been little discussion about the potentials included in networks pre-trained for sound recognition, which are inherently suited for time-frequency tasks. This work argues that deep architectures trained for music recognition and sound detection can perform machine diagnosis. The YAMNet convolutional network was designed to serve extremely efficient mobile applications for sound detection, and it was originally trained on millions of data extracted from YouTube clips. That framework is employed to detect bearing faults for the CWRU dataset. It is shown that transferring knowledge from sound and music recognition to bearing fault detection is successful. The maximum accuracy is achieved using a few hundred data for fine-tuning the fault diagnosis model.

The training of Artificial Intelligence algorithms for machine diagnosis often requires a huge amount of data, which is scarcely available in industry. This work shows that convolutional networks pre-trained for audio classification already contain knowledge for classifying bearing vibrations, since both tasks share the need to extract features from spectrograms. Knowledge transfer is realized through transfer learning to identify localized defects in rolling element bearings. This technique provides a tool to transfer the knowledge embedded in neural networks pre-trained for fulfilling similar tasks to diagnostic scenarios, significantly limiting the amount of data needed for fine-tuning. The VGGish model was fine-tuned for the specific diagnostic task by handling vibration samples. Data were extracted from the test bench for medium-size bearings specially set up in the mechanical engineering laboratories of the Politecnico di Torino. The experiment involved three damage classes. Results show that the model pre-trained using sound spectrograms can be successfully employed for classifying the bearing state through vibration spectrograms. The effectiveness of the model is assessed through comparisons with the existing literature.

Recent advances in Large Language Models (LLMs) enable agentic systems that combine perception, reasoning, and action across the enitre Predictive Maintenance (PdM) lifecycle, including machine fault diagnosis. However, the literature on LLM-driven agents for PdM remains fragmented and lacks a unified view on contemporary frameworks such as Model Context Procotol. This paper reviews discriminative, generative, and LLM-based approaches for PdM and consolidates fragmented evidence on LLM-driven AI agents. Namely, it introduces agentic AI concepts for PdM and develops an analysis of potential applications, challenges, and risks in light of agency theory, while mapping drivers and barriers to adoption based on recent evidence from industry analysis. Findings indicate near-term value for information and decision-support agents, while higher autonomy needs stronger governance, benchmarks, and safety evidence.

Predictive maintenance strategies are established in the industrial context on account of their benefits in terms of costs abatement and machine failures reduction. Among the available techniques, vibration-based condition monitoring (VBCM) has notably been applied in many bearing fault detection problems. The health indicators construction is a central issue for VBCM, since these features provide the necessary information to assess the current machine condition. However, the relation between vibration data and its sources intimately related to bearing damage is not effortlessly definable from a diagnostic perspective. This study discloses a diagnostic investigation performed both on the vibration signal and on the contact pressure signal that is supposed to be one of main forcing terms in the dynamic equilibrium of the damaged bearing. Envelope analysis and spectral kurtosis (SK) are applied to extract and compare diagnostic features from both signals, referring to the Case Western Reserve University (CWRU) case-study. Namely, health indicators are constructed by means of physical considerations based on the effect of faults on the signal power contents. These indicators show to be promising not only for damage detection but, also, for damage severity assessment. Moreover, they provide an invaluable reading key of the link occurring between the contact pressure path and the vibration response.

The Intelligent Fault Diagnosis of rotating machinery calls for a substantial amount of training data, posing challenges in acquiring such data for damaged industrial machinery. This paper presents a novel approach for generating synthetic data using a Generative Adversarial Network (GAN) with cycle consistency loss function known as cycleGAN. The proposed method aims to generate synthetic data that could effectively replace real experimental data. The generative model is trained to transform wavelet images of simulated vibrational signals into authentic data obtained from machinery with damaged bearings. The utilization of Maximum Mean Discrepancy (MMD) and Fréchet Inception Distance (FID) demonstrates a noteworthy resemblance between synthetic and real experimental data. Also, the generative model enables the synthesis of data that may have been entirely lacking from the experimental observation, indicating generative zero-shot learning capabilities. The efficacy of synthetic data in training diagnosis algorithms by means of Transfer Learning (TL) on Convolutional Neural Networks (CNNs) has been demonstrated to be comparable to that of real data. The study has been validated by means of the test rig for medium-sized industrial bearings accessible at the Politecnico di Torino.