Thesis Format
Integrated Article
Degree
Doctor of Philosophy
Program
Electrical and Computer Engineering
Collaborative Specialization
Artificial Intelligence
Supervisor
Shami, Abdallah
Abstract
The advancement of Internet of Things (IoT) devices in computational and storage capabilities enabled their use in data- and resource-intensive applications, including Machine Learning (ML), across various disciplines. This thesis focuses on the challenges of building an ML pipeline in industrial settings, particularly in the stages of data acquisition, data processing, model training, and model interpretation. These challenges involve data privacy, data scarcity, model construction optimization, distributed learning, interpretability of Deep Learning (DL) results, and scarcity of communication and computational resources. To address these challenges, this thesis proposes solutions aimed at facilitating the deployment of ML pipelines in industrial settings. Under the broad theme of building capable ML models, this thesis proposes two methods for feature engineering of time-series data, capturing their underlying dynamics, and incorporating low-complexity ML algorithms and transfer learning to mitigate data scarcity. Additionally, a novel hyper-parameter optimization method is devised, combining reinforcement learning (RL) and transformer architecture, to expedite and enhance its transparency. Following this, an extensive analysis is conducted on integrating data-driven techniques to optimize warehouse energy consumption. Based on data collected from real-world environments, a use-case study evaluates different feature engineering techniques and the physical phenomena connected to Heating, Ventilation, and Air Conditioning (HVAC) operations encountered in enclosed spaces. Recognizing the importance of DL model explainability in aiding decision-making processes, especially, in high-stakes industrial environments, this thesis addresses two use cases: identifying features affected by industrial noise and detecting drifting features caused by cyber-security attacks. Furthermore, the thesis tackles the challenges of communication resource scarcity in the Federated Learning (FL) paradigm by proposing quantization techniques and investigating model weight similarity in heterogeneous FL scenarios. These contributions, available via open-source software, are poised to significantly advance the deployment and effectiveness of ML pipelines in industrial environments.
Summary for Lay Audience
The development of smart devices, referred to as the Internet of Things (IoT) enabled the utilization of advanced technology like Machine Learning (ML) that depends on large amounts of collected data to infer inherent patterns and help in automation decisions. However, building effective ML systems for industrial applications is hampered by many challenges. This thesis explores these challenges and proposes solutions to enhance the applicability of ML in real-world implementations. Applying transformation strategies for the raw data is essential for ML models to uncover hidden patterns. This also extends to cases when data is scarce, whereby both scenarios are encountered in an industrial environment. This thesis proposes novel methods that enhance data processing so that models can learn more effectively and frameworks that facilitate these models' portability to different settings. This research also investigates the potential of ML in improving energy efficiency in warehouses by exploring the physical phenomenon in warehouses and real-world data. A key focus of this thesis is addressing the understandability aspect of the decision-making process of deep learning, which is essential for making informed decisions in manufacturing and cyber-security disciplines. Moreover, this thesis also addresses improving the ML system function in scenarios where communication resources are limited and data privacy needs to be preserved. The proposed solutions are publicly available as open-source software so that they can be integrated into industrial applications. In summary, these contributions improve the efficiency of ML systems to realize the promise of smart technologies.
Recommended Citation
Shaer, Ibrahim, "Applications of Machine Learning in Industrial Buildings" (2024). Electronic Thesis and Dissertation Repository. 10675.
https://ir.lib.uwo.ca/etd/10675
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Included in
Computer and Systems Architecture Commons, Other Electrical and Computer Engineering Commons