http://lutpub.lut.fi:80/handle/10024/158302 2026-09-17T00:35:16Z 2026-09-17T00:35:16Z Misal, Noopoor http://lutpub.lut.fi:80/handle/10024/172938 2026-09-11T09:30:11Z 2026-10-09T00:00:00Z

The idea of a better world as a click away : exploring digital platforms as enablers of user behaviour for social sustainability Misal, Noopoor Sustainable development has gained prominence in policy and academic discourse, but progress toward its objectives remains uneven. The social pillar of sustainability, which covers equity, safety, participation, and community well-being, has received less attention than the environmental pillar. Sustainability is enacted not by structural change alone but through the everyday behaviours of individuals who engage with the innovations designed to advance it. This dissertation examines how digital platforms shape user behaviours and experiences related to sustainability, with a particular focus on the social dimension. The dissertation comprises four studies. The first systematic review examines how digital platforms contribute to achieving the Sustainable Development Goals. The second and third examine socio-cognitive and design factors shaping socially sustainable behaviour on a digital platform, integrating the Theory of Planned Behaviour, the Technology Acceptance Model, and Equity Theory; the third also situates platforms within the Multi-Level Perspective in sustainability transitions. The second uses ridge regression and the third uses moderated regression and mediation analysis, both drawing on survey data from a social sustainability platform in the Global South. The fourth adopts a critical lens grounded in techno-solutionism, using sentiment analysis and neural topic modelling on user discourse from four established sustainability-oriented platforms in the Global North. The findings show that platforms cultivate socially sustainable behaviour through distinct mechanisms: self-efficacy, perceived ease of use, effort fairness, internalised social concern, and users’ variety-seeking behaviour. User discourse reveals that sustainability is rarely foregrounded in everyday engagement with platforms designed to advance it. The dissertation contributes to platform scholarship by centring the social dimension and by shifting attention toward user discourse through the critical lens of techno-solutionism, and to sustainability transitions research by providing behavioural micro-foundations. It concludes that platforms advance sustainability only when design and engagement conditions translate user participation into substantive outcomes and recommends attending to these conditions rather than treating adoption as evidence of impact.

2026-10-09T00:00:00Z Schneider, Fabian http://lutpub.lut.fi:80/handle/10024/172904 2026-09-04T12:00:17Z 2026-09-21T00:00:00Z

Score-based diffusion models in Bayesian inverse problems Schneider, Fabian This thesis investigates the use of score-based diffusion models for Bayesian inverse problems, with a focus on high-dimensional imaging applications. After introducing relevant applications such as inpainting, computed tomography (CT), deblurring and diffuse optical tomography (DOT) and reviewing the mathematical framework of Bayesian inverse problems, the thesis surveys recent developments in diffusion-based generative modelling, including conditional and unconditional score functions, infinite-dimensional formulations, and neural operator-based approximations. This work develops a principled framework for interpolating between two score functions, such as a data-driven and a model-based score. It is demonstrated that, in the small diffusion-time regime, the proposed interpolation approximates the score of a geometric mixture distribution. A tunable mixing parameter enables the adaptive weighting of the two score components without retraining, and the theoretical findings are supported by numerical experiments in Gaussian settings. This dissertation introduces UCoS, a formulation expressing conditional score functions as affine transformations of intermediate task-dependent scores for linear inverse problems. The approach is derived in the infinite-dimensional Hilbert-space setting and leads to a novel framework for posterior sampling that avoids repeated evaluations of the forward operator during inference. A modified score matching objective is introduced to enable training, and numerical errors are theoretically quantified. Numerical experiments demonstrate the performance of the proposed methods across several imaging problems. In inpainting with Gaussian priors, the proposed and conditional methods accurately reproduce posterior statistics, while unconditional approaches exhibit varying deviations. In large-scale CT imaging, UCoS achieves significant computational speed-ups and improved performance in low-parameter regimes. In high-dimensional deblurring, UCoS produces more realistic reconstructions than competing methods while requiring less computation. In a DOT problem with experimental and simulated data, the proposed methods show strong performance, with regularised variants improving reconstruction quality and coverage in challenging settings. Overall, the results highlight the potential of score-based diffusion models for scalable Bayesian inference and demonstrate how learned generative models can be integrated with a classical inverse problem structure. The thesis further suggests that hybrid and task-dependent approaches offer promising directions for improving both computational efficiency and reconstruction quality in large-scale scientific imaging problems.

2026-09-21T00:00:00Z Zhang, Li http://lutpub.lut.fi:80/handle/10024/172901 2026-09-04T11:30:19Z 2026-09-25T00:00:00Z

A coordinate partitioning framework with iterative refinement for coupled multibody dynamics and hydraulic systems Zhang, Li Real-time simulation of constrained multibody systems is challenging due to complex constraint networks and strong nonlinearities arising from coupled domains. In coordinate partitioning formulations, updating dependent coordinates is often the dominant computational cost and significantly affects numerical robustness and accuracy. This thesis develops and evaluates efficient numerical strategies for dependent coordinate updates and extends them to real-time simulation of nonlinear hydraulically coupled multibody systems. A unified framework is established to assess three update strategies: exact update, Newton increment, and Newton-Raphson, applied to planar and spatial benchmark mechanisms. Profiling indicates that dependent coordinate updates account for 60–90% of the total runtime, motivating targeted optimization. To improve scalability, a global coordinate partitioning approach with iterative refinement is introduced. Numerical results show that it achieves accuracy comparable to relative-coordinate semi-recursive formulations while significantly reducing computational cost. Finally, a coupled multibody hydraulic formulation is proposed. Iterative refinement is applied to both independent coordinate integration and constraint enforcement, avoiding conventional nonlinear solvers. The results demonstrate reduced computational effort while maintaining accurate constraint satisfaction and energy-consistent behavior. Overall, the thesis provides practical guidelines for selecting update strategies under real-time and high accuracy requirements, and introduces refinement-based techniques that enhance efficiency and robustness in multi-physics multibody simulations.

2026-09-25T00:00:00Z Bakhaev, Stepan http://lutpub.lut.fi:80/handle/10024/172656 2026-08-06T07:10:01Z 2026-08-21T00:00:00Z

On stakeholder dialectics in digital identity infrastructures Bakhaev, Stepan This thesis studies change in identity management in the context of digital public service delivery. Identity management involves actors, processes and supporting technologies for the creation, management and use of digital identities—the information about individuals necessary for access to varied services. Software-based systems for identity management are undergoing a considerable change, coupled with the decentralised communication and distributed decision-making for digital identities, which gives individuals greater control over personal information. This brings organisational complexity in system development and calls for consideration of stakeholder relations when specifying system requirements, managing interests and making architectural decisions that influence the scope of change in the digital infrastructures of public services. The goal of this thesis is to analyse stakeholder relations during system development. The analysis is based on a series of qualitative field studies within a large inter-organisational project in Europe. The project focused on the development of an electronic identification system for online public services and involved software developers, citizens and public administrations in a collaborative design process. The research combined observational and participatory components from the empirical setting and proceeded in three phases. First, the study explored requirements for the system design using co-creation principles. Second, the study investigated architectural assumptions underpinning design decisions during the system development. Third, the study conducted a comprehensive stakeholder analysis to examine the role of symbolic properties and trust in the system adoption. The results emphasise the need to shift attention in identity management decentralisation from technical designs to the social and organisational factors that influence design during system development. The integration of user-centred identity management systems into the digital infrastructures of public services requires engaging stakeholders in a dialogic exploration of their infrastructural relations. By supporting institutional transparency and inclusivity in the design of digital identity technology, practitioners and policymakers can promote trust in emergent decentralised solutions. Thus, this thesis aims to contribute to knowledge of identity management for public services as socio-technical systems shaped by stakeholders’ infrastructural relations.

2026-08-21T00:00:00Z