https://bura.brunel.ac.uk/handle/2438/210 2026-08-23T17:29:47Z https://bura.brunel.ac.uk/handle/2438/33697 Title: Intelligent outlier detection under imperfect industrial data conditions Authors: Fang, Jingzhong Abstract: With the growing complexity and data intensity of modern industrial systems, intelligent data analysis (IDA) has become essential for reliable data interpretation and efficient operation. By leveraging advanced analytical and computational techniques, IDA can handle complex, largescale data sets, thereby providing valuable insights for industrial operations and significantly improving the stability and efficiency of industrial processes. Nevertheless, data quality remains a fundamental prerequisite for IDA performance, as it directly affects the accuracy and trustworthiness of derived insights. Due to complex operating conditions, high cost, and limited availability of expert annotation, high-quality data is often difficult to acquire. The imperfections in the data can affect training and evaluation, thereby reducing the overall performance in real-world applications. In practice, imperfections in industrial data generally fall into two main categories: 1) data scarcity, such as class imbalance or limited sample size; and 2) label quality issues, such as missing labels or noisy labels. In this thesis, we deal with the above-mentioned data imperfections arising from the data acquisition process and the annotation process to develop innovative approaches for robust and reliable IDA in industrial applications. It should be pointed out that all approaches developed in this thesis have been evaluated and applied to outlier detection on imperfect industrial data collected from real-world wire arc additive manufacturing (WAAM) processes. • To achieve outlier detection on unlabeled data, an improved optimization-based clustering algorithm is proposed, where the initial locations of the cluster centroids in the fuzzy C-means algorithm are optimized by an improved particle swarm optimization (PSO) algorithm. An adaptive switching randomly perturbed particle swarm optimization (ASRPPSO) algorithm is developed to enhance the convergence rate and particle’s search ability of the PSO algorithm, where a distance-based weighting strategy and switching strategy are introduced to update parameters of the optimizer. The particles can conduct a thorough search by accounting for both evolutionary states and the distances to the global and personal best positions, thereby improving the convergence rate and solution accuracy. Via the ASRPPSO-based selection of optimal clustering centroids, the proposed algorithm does not rely on centroid initialization, thereby facilitating a better cluster partition. • With the aim to guarantee the outlier detection performance on imbalanced data under the small sample problem, an optimized deep transfer learning framework is developed, where a novel deep domain adaptation strategy is designed to minimize the cross-domain discrepancies, the weighting factors are designed to handle the data imbalance problem, and the PSO algorithm is utilized for hyper-parameters tuning. By leveraging the domain knowledge and optimal hyper-parameter selection, the developed framework effectively balances performance and efficiency when dealing with outlier detection tasks on imbalanced data under the small sample problem. • To handle the noisy label problem under limited-data conditions in outlier detection, a novel Transformer-embedded learning with noisy labels framework with fuzzyclustering- assisted contrastive learning (TFCCL) is developed, where a fuzzy-clusteringassisted contrastive learning approach, a dynamic two-stage training scheme and a joint learning strategy are introduced to train the outlier detector. The TFCCL framework integrates supervised learning with contrastive learning, thereby reducing reliance on potentially noisy labels and enhancing model robustness. • For the purpose of outlier detection under the noisy label problem when sufficient data are available, a role-differentiated learning with noisy labels (RD-LNL) approach is put forward, where a leader-follower-inspired sample selection (LFSS) strategy is proposed for identifying potential clean samples for model training. A selection metric and an adaptive selection scheme are designed to adjust the number of selected samples. By combining the joint learning and adaptive sample selection, the RD-LNL framework achieves robust outlier detection in the presence of noisy labels. • To fully evaluate the application potential of the developed frameworks. All the developed frameworks are applied to outlier detection tasks on WAAM data sets collected from real-world manufacturing processes to examine their adaptability, robustness, and generalization capability. The experimental results demonstrate the effective and robust performance of the developed outlier detection frameworks on WAAM data sets across various data imperfection conditions. Description: This thesis was submitted for the award of Doctor of Philosophy and was awarded by Brunel University London 2026-01-01T00:00:00Z https://bura.brunel.ac.uk/handle/2438/33501 Title: Smart cities in Qatar toward a sustainable digital transformation Authors: Majareh, Maryam Abstract: The rapid progression of digital transformation has reshaped urban landscapes globally, fostering the development of smart cities. This thesis investigates the sustainability of digital transformation within the context of smart cities, focusing specifically on Qatar. With its ambitious national vision and substantial investments in smart city technologies, Qatar is a critical case study for understanding how digital transformation can be implemented to drive long-term sustainability. The research delves into critical areas such as governance frameworks, technological infrastructure, data management, and citizen engagement, all crucial for building resilient and sustainable smart cities. It also explores how digital solutions contribute to environmental, social, and economic sustainability, highlighting the role of emerging technologies like AI, IoT, and blockchain in improving city operations, resource management, and quality of life. The study uses quantitative and qualitative data from a survey and interviews with expert stakeholders to evaluate digital transformation efforts in Qatar's smart cities, including Lusail and Msheireb Downtown. It assesses their alignment with sustainability goals and recommends fostering an inclusive, efficient, and future-proof digital infrastructure. It also identifies challenges related to cybersecurity, regulatory frameworks, and digital inclusion, offering insights on how these can be mitigated to ensure the sustainability of Qatar's smart city initiatives. This research identifies key lessons applicable to other developing smart cities through evaluating Qatar’s approach, including the requirement of a well-defined regulatory framework, which can help in maintaining a balance between innovation and security, the significance of public-private partnerships in increasing technological adoption, and the role of digital literacy programs in ensuring inclusive participation. The thematic map and PRISMA synthesis together show that long-term sustainability depends not only on infrastructure investment, but also on governance capacity, interoperability, stakeholder trust, and adaptive regulation. This research contributes to the broader discourse on smart cities by presenting a comprehensive framework for sustainable digital transformation that is supported by survey analysis, stakeholder interviews, thematic mapping, and PRISMA-based evidence synthesis. It offers valuable lessons for Qatar's policymakers, technologists, urban planners, and others. Description: This thesis was submitted for the award of Doctor of Philosophy and was awarded by Brunel University London 2025-01-01T00:00:00Z https://bura.brunel.ac.uk/handle/2438/33430 Title: Understanding the search space: Investigations into the nature of software modularisation Authors: Mann, Ashley J. Abstract: A relationship exists between the functionality of software systems and their complex-ity. As the number of features implemented increases, the systems complexity also grows, accompanied by the expansion of the number of artefacts and their intricate in-terrelationships [1]. Description: This thesis was submitted for the award of Doctor of Philosophy and was awarded by Brunel University London 2026-01-01T00:00:00Z https://bura.brunel.ac.uk/handle/2438/33356 Title: Towards an Adaptable Architecture for Digital Twin (AADT) Authors: Nwogu, Chukwudi Abstract: A Digital twin (DT) is a virtual replica of a physical object, which has the capability of integrating with a virtual object, such that they exchange data and use the data exchanged to improve each other. It is a confluence of Industry 4.0 technological innovations, such as big data, artificial intelligence, modelling and simulation, Internet of things, optimisation techniques, cyber-physical systems, amongst others; the blend of these technologies fused together for the development of a DT is driven by its use case. A DT has enormous potential, and there is a consensus among the industry, academia and governments that it is one of the most pivotal technologies in Industry 4.0 era that will play a major role in shaping the society. This widely acknowledged notion about DT has not translated into a joint effort to standardise it. As a result, there exists neither a generally accepted definition nor architecture for digital twin. The lack of standard definition, framework or architecture for DT may have a negative impact on the wider adoption and development of DT. The state-of-the-art in DT architecture, for instance, reveals that most of the architectures are designed for specific domains and/or technologies and have components that are named in such a manner that it is difficult to identify commonality in purpose and functionalities. To contribute to the taxonomy of DT architectural components, this study proposes an adaptable architecture for DT (AADT), which is developed based on design science research (DSR) principles. AADT directly addresses the architectural chaos inherited from the growth era of the digital architectural development by establishing standard components traceable to digital twin definitions, requirements, and mandatory functionalities. Rather than proposing yet another domain-specific architecture, AADT provides a systematic process for deriving architectures from requirements, enabling consistent yet flexible implementations. To support the implementation of digital twins, this research develops an implementation framework for digital twin (IFDT) that is a confluence of software development lifecycle and principles of project controls. IFDT consists of a stage gate within every lifecycle phase, which ensures that a digital twin development project is subject to business case viability test and stakeholders’ approval as it progresses from one lifecycle phase to another. AADT is evaluated with the guidance of framework for evaluating design science (FEDS) and ISO Standard 9126. The evaluation of AADT results in the development of a taxonomy of architectural adaptiveness for digital twin systems. The taxonomy organises structural, behavioural, functional and quality adaptiveness into a coherent analytical framework; and therefore, serves as both a conceptual contribution and a practical guide for evaluating future digital twin architectures. In summary, this research, contributes to the taxonomy of digital twin architectural components; develops a technology-agnostic digital twin architecture that can adapt to the requirements of disparate use cases from a wide range of domains; proposes an implementation framework for digital twin; and develops a taxonomy of architectural adaptiveness of digital twin systems. Description: This thesis was submitted for the award of Doctor of Philosophy and was awarded by Brunel University London 2025-01-01T00:00:00Z