Predictive maintenance plays a major role in operational cost reduction in several industries and the railway industry is no exception. Predictive maintenance relies on real time data to predict and diagnose technical failures. Sensor data is usually utilized for this purpose, however it might not always be available. Events data are a potential substitute as a source of information which could be used to diagnose and predict failures. This thesis investigates the use of events data in the railway industry for failure diagnosis and prediction. The proposed approach turns this problem into a sequence classification task, where the data is transformed into a set of sequences which are used to train the machine learning algorithm. Long Short-Term Memory neural network is used as it has been successfully used in the past for sequence classification tasks. The prediction model is able to achieve high training accuracy, but it is at the moment unable to generalize the patterns and apply them on new sets of data. At the end of the thesis, the approach is evaluated and future steps are proposed to improve failure diagnosis and prediction.

This master thesis concerns modern method of TPM used in the production companies for maintenance of machinery. In the theoretical part, general maintenance is first described, as well as the TPM method. There is also described the technical diagnostics, which was emphasized in the practical part in terms of the use of vibrodiagnostics in predictive maintenance. In the practical part there is presented the proposal for introduction of the TPM method at the SMC Industrial Automation s.r.o. in Vyškov.

The master’s thesis is focused on online diagnostics of the machining process. In the theoretical part are presented maintenance possibilities of machine tools. A whole chapter is devoted to vibrodiagnostics, which describes vibration sensors, their attachment to the measured object and methods of vibration signal processing. The practical part lies in creating a software diagnostic solution for chosen PLC and sensors. The functionality of the proposed system was verified during experimental machining, which also provided valuable data for the correct setting of the system.

The Grate-Kiln-Cooler process is a commonly used method of sintering during iron ore pelletization, where the pellets are formed, dried, and hardened. The pellets are oxidized in the rotating Kiln, turning magnetite (Fe3O4) to hematite (Fe2O3), making the pellets attain suitable metallurgical attributes for further processing. The process is constantly exposed to thermal and mechanical stress, causing equipment degradation and thus unwanted production stops due to internal process disturbances. A suitable maintenance policy is required to cope with the risk of equipment degradation causing these production stops. Predictive maintenance (PdM) is the most current maintenance policy, utilizing a substantial amount of production data to foresee breakdowns and thus indicating the need for maintenance efforts to prevent them from occurring.           The global supplier of iron ore products, Loussavaara-Kiirunavaara Aktiebolag (LKAB), operates three pelletization plants in Kiruna. One of these pelletization plants experiences availability below desired levels. This hampers the plant from fulfilling its yearly production goals, resulting in lost revenue. This master's thesis aimed to increase the understanding of which causes influence the Grate-Kiln-Cooler process' availability. When these causes were identified, the aim was to develop a method of monitoring these to predict the need for maintenance (i.e., incorporating a PdM policy) to mitigate the risk of production stops. The work has been conducted by utilizing the systematic problem-solving DMAIC methodology.    The refractory material was identified as the primary contributor to the low availability in the investigated plant. Using principal component analysis (PCA) and statistical process control (SPC), a Hotelling T2 chart based on principal components was established to monitor the refractory material's condition. In this context, the combined usage of PCA and SPC highlighted three possible tendencies in the Kiln that potentially damaged the refractory material, causing production stops. The observed tendencies with the possibility of damaging the refractory material were; abnormally high refractory material temperatures, periods where the pellets' temperature exceeded the refractory material's temperature, and sporadic heat fluctuations in the refractory material.  The utilized Hotelling T2 chart provided a current state evaluation of the refractory material's condition and thus indicated the need for maintenance efforts. However, it was impossible to predict breakdowns by identifying patterns in either the T2-statistics or the individual charts. The inability to predict stops was derived from obstacles related to lacking documentation, deficient data, and that the time for breakdown is difficult to determine accurately. These obstacles hinder the prediction of breakdowns and, therefore, need to be dealt with to facilitate the implementation of a successful PdM strategy.

Recent refinements in Industry 4.0 and Machine Learning demonstrate the positive effects of using deep learning models for intelligent maintenance. The primary benefit of Deep Learning (DL) is its capability to extract attributes and make fast, accurate, and automated predictions without supervision. However, DL requires high computational power, significant data preprocessing, and vast amounts of data to make accurate predictions for intelligent maintenance. Given the considerable obstacles, meta-learning has been developed as a novel way to overcome these challenges. As a learning technique, meta-learning aims to quickly acquire knowledge of new tasks using theminimal available data by learning through meta-knowledge. There has been less research in the area of using meta-learning for Predictive Maintenance (PdM) and we considered it necessary to conduct this review to understand the applicability of meta-learning’s capabilities and functions to PdM since the outcomes of this technique seem to be rather promising. The review started with the development of a methodology and four research questions: (1) What is the taxonomy of meta-learning for PdM?, (2) What are the current state-of-the-art methodologies? (3) Which datasets are available for meta-learning in PdM?, and (4) What are the open issues, challenges, and opportunities of meta-learning in PdM?. To answer the first and second questions, a new taxonomy was proposed and meta-learnings role in predictive maintenance was identified from selected 55 papers. To answer the third question, we determined which types of datasets and their characteristics exist for this domain. Finally, the challenges, open issues, and opportunities of meta-learning in predictive maintenance were examined to answer the final question. The results of the research questions provided suggestions for future research topics.

Surface treatments are often used in the manufacturing industry to change the surface of a product, including its related properties and functions. The occurrence of degradation and corrosion in surface treatment processes can lead to critical breakdowns over time. Critical breakdowns may impair the properties of the products and shorten their service life, which causes increased lead times or additional costs in the form of rework or scrapping.  Prevention of critical breakdowns due to machine component failure requires a carefully selected maintenance policy. Predictive maintenance is used to anticipate equipment failures to allow for maintenance scheduling before component failure. Developing predictive maintenance policies for surface treatment processes is problematic due to the vast number of attributes to consider in modern surface treatment processes. The emergence of smart sensors and big data has led companies to pursue predictive maintenance. A company that strives for predictive maintenance of its surface treatment processes is Seco Tools in Fagersta. The purpose of this master's thesis has been to investigate the occurrence of critical breakdowns and failures in the machine components of the chemical vapor deposition and post-treatment wet blasting processes by mapping the interaction between its respective process variables and their impact on critical breakdowns. The work has been conducted as a Six Sigma project utilizing the problem-solving methodology DMAIC.  Critical breakdowns were investigated combining principal component analysis (PCA), computational fluid dynamics (CFD), and statistical process control (SPC) to create an understanding of the failures in both processes. For both processes, two predictive solutions were created: one short-term solution utilizing existing dashboards and one long-term solution utilizing a PCA model and an Orthogonal Partial Least Squares (OPLS) regression model for batch statistical process control (BSPC). The short-term solutions were verified and implemented during the master's thesis at Seco Tools. Recommendations were given for future implementation of the long-term solutions. In this thesis, insights are shared regarding the applicability of OPLS and Partial Least Squares (PLS) regression models for batch monitoring of the CVD process. We also demonstrate that the prediction of a certain critical breakdown, clogging of the aluminum generator in the CVD process, can be accomplished through the use of SPC. For the wet blasting process, a PCA methodology is suggested to be effective for visualizing breakdowns.

As part of Volvo Penta's initiative to further the development of predictive maintenance in their field test environments, this thesis compares neural networks in an effort to predict the occurrence of three common diagnostics trouble codes using field test data. To quantify the neural networks' performances for comparison a number of evaluation metrics were used. By training a multitude of differently configured feedforward neural networks with the processed field test data and evaluating the resulting models, it was found that the resulting models perform better than that of a baseline classifier. As such it is possible to use Volvo Penta's field test data along with neural networks to achieve predictive maintenance. It was also found that Long Short-Term Memory (LSTM) networks with methodically selected hyperparameters were able to predict the diagnostic trouble codes with the greatest performance among all the tested neural networks.

The success of Internet of Things solutions allowed the establishment of new applications such as smart hydroponic agriculture. One typical problem in such an application is the rapid degradation of the deployed sensors. Traditionally, this problem is resolved by frequent manual maintenance, which is considered to be ineffective and may harm the crops in the long run. The main purpose of this thesis was to propose a machine learning approach for automating the detection of sensor fault drifts. In addition, the solution’s operability was investigated in a cloud computing environment in terms of the response time. This thesis proposes a detection algorithm that utilizes RNN in predicting sensor drifts from time-series data streams. The detection algorithm was later named; Predictive Sliding Detection Window (PSDW) and consisted of both forecasting and classification models. Three different RNN algorithms, i.e., LSTM, CNN-LSTM, and GRU, were designed to predict sensor drifts using forecasting and classification techniques. The algorithms were compared against each other in terms of relevant accuracy metrics for forecasting and classification. The operability of the solution was investigated by developing a web server that hosted the PSDW algorithm on an AWS computing instance. The resulting forecasting and classification algorithms were able to make reasonably accurate predictions for this particular scenario. More specifically, the forecasting algorithms acquired relatively low RMSE values as ~0.6, while the classification algorithms obtained an average F1-score and accuracy of ~80% but with a high standard deviation. However, the response time was ~5700% slower during the simulation of the HTTP requests. The obtained results suggest the need for future investigations to improve the accuracy of the models and experiment with other computing paradigms for more reliable deployments.

Production of steel powder is done by atomization of a molten steel stream. Atomization is done by feeding high pressure nitrogen gas through nozzles, creating jets of gas which scatter the molten steel stream into powder. The steel powder falls through the atomization tower whilst it cools and solidifies. Finally, the steel powder is transported for further processing. The compressor is used for two main purposes, to compress the nitrogen gas to desired pressure and enable recycling of nitrogen gas. As nitrogen is inert and do not react with its surrounding, the gas can be recycled. Filtering nitrogen gas from the atomization process, one is able to reuse the gas, which is led to the inlet side of a compressor. A closed loop is thus created which is economically important. In 2021 a major compressor failure occurred, which caused large production losses. iba systems is a commercially available product extensively utilized in the Swedish steel industry for data acquisition, production monitoring and generating key performance indicators. Therefore, this thesis investigates what modules and functionality iba systems have to offer. Process and machine signals are studied to assess both their utility in predicting machine failure and relevant iba modules for the predictive maintenance purposes, based on a literature review. This thesis shows the possibility to implement an anomaly detection to detect abnormal behavior, related to historic compressor failure. Estimating when maintenance is needed is possible but requires implementation of new sensors to obtain useful information, mainly vibration data from machinery. Anomaly detection is implemented using ibaAnalyzer. Additional analysis is done in Matlab.

This dissertation aims to provide efficient process defect identification methods for advanced manufacturing environments using AI tools/algorithms with limited labeled data availability. Asset and equipment quality become highly sensitive in sustaining virtuous performance and safety in various manufacturing domains. Internally generated process imperfections degrade finished products' optimum performance and mechanical attributes. The evolution of big data and intelligent sensing systems leverage data-driven defect identification in advanced manufacturing environments. Widely adopted data-driven process anomaly detection methods assume that the training (source) and testing (target) data follow the same distribution and that labeled data are available in both source and target domains. However, the source and target sometimes follow different distributions in real-world manufacturing environments as the diversity of industrialization processes leads to heterogeneous data collection under different production conditions. Such a case significantly limits the performance of AI algorithms when distribution discrepancy exists. Moreover, labeling data is typically costly and time-consuming, signifying that identifying process defects is limited by rare labeled data. Also, more realistic industrial applications incorporate fewer defect data than ordinal data and unforeseen target defects, leading to complications in understanding the process behaviors in various aspects. Therefore, we introduced methodological principles, including unsupervised grouping, transfer learning, data augmentation, and ensemble learning to address these limitations in advanced operations. First, rapid porosity prediction methodology for additive manufacturing (AM) processes under varying process conditions is developed by leveraging knowledge transfer from existing process conditions. Second, designing an effective classification method concerning time series signals to advance predictive maintenance (PdM) for machine state prediction is discussed. Finally, a data augmentation-based stacking classifier approach is developed to enhance the precision of predicting porosity, even when limited porosity data is available.