[diss] Insinööritieteiden korkeakoulu / ENG

Permanent URI for this collectionhttps://aaltodoc.aalto.fi/handle/123456789/49

Browse

Recent Submissions

Now showing 1 - 20 of 750
  • Geothermal energy piles design, sizing and modelling
    (2025) Fadejev, Jevgeni
    School of Engineering | Doctoral thesis (article-based) | Defence date: 2025-11-20
    This thesis addresses the challenges associated with the design, sizing, and modelling of geothermal energy piles (GEPs), the lack of validated methods for their use as a renewable heating and cooling solution for nearly zero-energy buildings (NZEBs). GEPs provide both load-bearing and ground heat exchange functions, making them well-suited for use with ground source heat pumps (GSHPs). However, their designs have often relied on assumptions originating from borehole heat exchangers (BHEs), which differ considerably from GEPs in geometry, thermal boundary conditions, and placement, as the layout of GEPs is dictated by the building’s foundation plan. This research aimed to develop and validate a modelling method for assessing the performance of GEPs with thermal storage coupled with a detailed whole building simulation model for a parametric study. The method was developed in IDA ICE and validated using COMSOL Multiphysics and realworld measurement data. The research methodology combined a systematic literature review, model development, validation, and demonstration of the modelling method’s performance using an as-built calibrated model with measured performance data from a commercial NZEB in Finland for energy analysis. A parametric study was conducted to support the development of a tabulated GEPs sizing method for early-stage design, considering factors such as heat pump sizing power, pile spacing and depth, soil type, and the presence of a thermal storage. The findings confirmed that conventional BHE-based modelling approaches are unsuitable for GEP systems due to major differences in thermal boundary conditions, particularly the influence of building floor slabs on ground temperature distribution. The validated GEP modelling method, implemented in IDA ICE and verified with COMSOL simulations, accurately captured these effects and showed strong agreement with measured data from a monitored NZEB in Finland. The model calibration procedure revealed unexpected plant operation due to improper control algorithms, highlighting the importance of monitoring and logging systems in buildings with unconventional plant designs to ensure proper operation and maintain long-term efficiency. According to parametric study results, seasonal thermal storage demonstrated notable improvements in energy efficiency and enabled a reduction in required pile length by over 50% in a specific case. A tabulated GEP sizing guide was developed to support early-stage design, enabling engineers to estimate system configurations effectively without relying on complex simulation tools. The method demonstrated in this thesis can be extended to any climate region and building type.
  • Data-driven transformation in construction management – From artificial intelligence to network modeling
    (2025) Nyqvist, Roope
    School of Engineering | Doctoral thesis (article-based) | Defence date: 2025-11-18
    Digital technologies like artificial intelligence (AI) and digital platforms are transforming the construction industry towards data-driven management, offering pathways to address long-standing issues like inefficiency, cost overruns, and project delays while enhancing quality and sustainability. However, realizing this potential can benefit from integrating digital solutions with complementary knowledge-structuring innovations for a holistic approach. This dissertation tackles key knowledge gaps by exploring the integration of data-driven digital innovations and knowledge-structuring innovations within construction. It addresses the research pitfall of isolating technology innovations from business model development and ecosystem evolution, emphasizing the need for an interconnected perspective. Furthermore, the dissertation investigates the under-explored potentiality of practical applications of generative AI across construction management, and particularly focuses on evaluating construction project risk management (CPRM) capabilities of humans and AI, and the need for management innovations complementing digital tools. Employing a mixed-methods approach, this dissertation addresses five key questions: (1) What are the primary barriers, drivers, and their associated actions for construction industry companies to consider in managing digital transformation? (2) What are the implications of databased digital innovations on the companies' business models in the construction industry? (3) What are the potentials of generative AI to enhance construction project management? (4) Taking a particular use case, what are the capabilities of generative AI in CPRM compared to human professionals? (5) How can a network-based approach provide a knowledge-structuring innovation for CPRM, and how can such innovations also support digital data-driven innovations? Key findings reveal: (1) synthesized barriers, drivers, and actionable strategies for navigating digital transformation; (2) how AI-driven platforms can reshape business models and leverage data, despite challenges like operational integration complexities and data security; (3) that generative AI shows substantial potential across seven construction management areas, notably outperforming human experts in CPRM, though practical use still needs human oversight; (4) the uncertainty network modeling (UNM) method introduced complements digital innovation by providing a knowledge-structuring approach to visualizing and managing interconnected risks, enhancing stakeholder collaboration, improving risk management, and providing AI with explicit project data. These findings demonstrate how construction management can be advanced by integrating digital innovations with methods that formalize human expertise. The results establish that the potential of AI is best unlocked through a partnership with human-centered approaches that make tacit knowledge explicit for machine utilization. Specifically, this research synthesizes these findings into an integrated four-layer framework: (1) digital transformation strategies to guide high-level adoption by overcoming key barriers; (2) digital solutions for implementing practical AI tools and platforms; (3) an integration layer where digital tools and knowledge-structuring methods are combined to foster human-machine collaboration; and (4) knowledge-structuring solutions, like the UNM method, which provide the structured, human-validated data essential for this synergy. This framework provides a transferable model for managing the synergy between technological innovation and human expertise in construction.
  • Brittle fracture in weld microstructures
    (2025) Hytönen, Noora
    School of Engineering | G5 Artikkeliväitöskirja | Defence date: 2025-11-14
    This dissertation investigates the phenomenon of brittle fracture in weld microstructures, with focus on nuclear safety and material ageing during extended long-term operation (ELTO) of nuclear power plants (NPPs). The research emphasizes the role of understanding the microstructural features of welds to ensure the integrity and safety of nuclear reactor components as they age. The study highlights the significant effects of thermal ageing and neutron irradiation on the mechanical properties of low-alloy steel (LAS), in terms of maintaining safe operational margins in NPPs. The primary focus is on two critical weld components in a nuclear reactor: the welds in the reactor pressure vessel (RPV), as well as dissimilar metal welds (DMWs) in the pressurizer surge nozzle safe-end. The study on RPV welds is conducted on harvested material from a decommissioned boiling water reactor (BWR) NPP combined with representative high-dose surveillance material. The studied DMW is a representative mock-up of a repair weld used in a pressurized water reactor (PWR) with double-sided Alloy 52 buttering weld joining LAS pressure vessel and stainless steel piping. The brittle fracture initiation and crack propagation are investigated from a microstructural perspective. The methodology includes a comprehensive multiscale analysis of microstructural features through various characterization techniques, such as scanning electron microscopy and transmission electron microscopy, along with hardness and mechanical testing to evaluate the strength and toughness properties of the welds. Key findings indicate that the microstructural characteristics of welds significantly influence their mechanical properties and susceptibility to brittle fracture. Two main types of second-phase particles are identified to initiate a cleavage fracture in a LAS weld: non-metallic oxide inclusions and carbides. The type of fracture initiator varies between the welds and is dependent on the specimen type, yet no conclusion can be made if the thermal ageing and/or irradiation affect the cleavage fracture initiation, as the mechanical testing shows near negligible operation-induced embrittlement in the RPV welds and auspicious material properties for LTO. In DMWs, three main types of fusion boundary microstructures were found along the LAS-Alloy 52 interface: sharp fusion boundary, martensitic phase, and partially melted zone. The microstructures correlate with varying levels of dilution and local strength mismatch, significantly affecting the crack propagation path. The strength mismatch and carbide evolution are influenced by welding parameters, post-weld heat treatment, and LTO temperature. The dissertation highlights that understanding of the metallurgical features and microstructural evolution of welds can lead to improved assessment of material performance and safety during ELTO. Current ageing management predictions are conservative, but to safely extend the operating lifetime, small details in embrittlement trend curves can increase the valuable time of safe operation. When performing testing of the weld metal, especially with a notched or pre-cracked sample, the location of the crack tip and the process zone must be known in terms of the local weld microstructure as it determines the local fracturing resistance.
  • Flavors of ventures, layers of context: Actions of small Finnish food and beverage ventures and how they change their landscapes
    (2025) Talvinko, Maria
    School of Engineering | Doctoral thesis (article-based) | Defence date: 2025-11-07
    Recent developments in entrepreneurship research have increasingly emphasized the importance of contextualization and the interdependence between entrepreneurial action and the external environment. This dissertation contributes to this ongoing shift by offering four distinct yet interconnected perspectives on the dynamics between ventures and their environments, grounded in the context of small Finnish food and beverage (F&B) ventures. This sector offers a vibrant setting, characterized by rapidly shifting consumer trends, technological advancements, global disruptions, and gradually evolving regulatory and cultural landscapes. Drawing on a qualitative study of 54 heterogeneous ventures engaged in forms of ‘everyday’ entrepreneurship—often overlooked in research that prioritizes high-growth, tech-driven startups—this dissertation reveals the coexistence of ‘standard model’ features such as innovation and growth orientation, alongside more relational, artisanal, and collaborative approaches to entrepreneurship. These ventures employ entrepreneurial actions that resemble a versatile toolkit used to address challenges, explore opportunities, and co-create with stakeholders in a complex and evolving landscape. The four articles comprising this dissertation each explore a specific dimension of the context–venture relationship. Article 1 examines experimentation and business model innovation during the COVID-19 crisis, highlighting its effects on both individual ventures and their communities. Article 2 examines how crisis-induced external enablers facilitated the development of opportunities among small businesses in Finland and California, highlighting the recursive creation of enabling conditions. Article 3 extends the dominant rational view of experimentation by identifying rational, social, and emergent forms driven by both external and internal factors. Finally, Article 4 combines stakeholder theory and the resource-based view to examine strategic stakeholder interactions, revealing a reliance on informal contracting and a relationally embedded F&B ecosystem. Overall, the thesis emphasizes the importance of ongoing research in the field of entrepreneurship, encompassing a broader range of venture types and a deeper consideration of the complex dynamics of context. The findings suggest opportunities for entrepreneurs, policymakers, and educators to broaden their perspectives and better include more diverse ventures and entrepreneurs.
  • Modelling, analysing and optimising ship energy systems
    (2025) Elg, Mia
    School of Engineering | Doctoral thesis (article-based) | Defence date: 2025-10-24
    Ship design and operation today focus largely on reducing the environmental impact of shipping. This is supported by the regulatory framework with an ambition to steer shipping towards net-zero greenhouse gas emissions. The current focus is on regulating ship carbon intensity and energy efficiency. Ship decarbonisation has a holistic impact on the entire ship design, and ship machinery and energy systems are at the heart of the change. This thesis explores ship energy system development towards decarbonised shipping by focusing on ship energy system modelling, analysis and optimisation through case examples. The context of the study lies in the methodology and questions that are relevant to early-stage ship design, where key decisions about ships and their systems are made. Optimisation was studied for two different ship design purposes: machinery selection and exploring the optimal set-up of waste heat recovery processes and battery system dimensioning. While selecting the main machinery topology for a cargo ship and a passenger ship, linear and non-linear optimisation were compared. Although this relatively straightforward task might favor the fast and efficient linear optimisation approach, in practice, machinery selection involves more dimensions than merely optimising fuel consumption. Therefore, as a secondary target, multi-objective optimisation was explored. This enabled comparison of the trade-offs between achieving the highest reductions in fuel consumption and the space required. While the machinery study involved defining a special algorithm for the process, a cruise ship energy system study took an existing ship energy simulation model as a starting point for optimisation. The objectives for the optimisation were fuel saving, investment cost, and engine running hours. The largest energy and fuel saving potential was recorded at close to 4% with the studied variables. Furthermore, the optimisation framework and result visualisation allowed for the examination of interesting sensitivities and relationships between the design variables and optimisation targets. The largest reduction in ship carbon emissions and energy consumption was witnessed in a conceptual case study of a cruise ship operating on a Mediterranean route. The application of several technologies, such as ultrasound antifouling, shore power, battery hybrid machinery, waste heat recovery and air lubrication led to combined fuel savings of 18,7% for the selected operational profile. By combining these technologies into a machinery powered by hydrogen, the total ship energy consumption was reduced by 25%. Synergies in ship energy system, such as machinery efficiency and ship heating requirements, contributed to this result. Nevertheless, even further development potential for the Mediterranean cruise ship was identified through the heat utilisation efficiency analysis. The heat system analysis relied on entropy generation calculation for selected ship heat systems, especially those related to waste heat recovery processes. Energy system modelling and simulation methods are an important part of ship design today, as they also help validate the rule compliance in the early design stages, where improvements can still be made to the ship at moderate costs. Optimisation methods enable an expansion of the solution exploration area in practical design tasks. Furthermore, integrating the optimisation into ship energy modelling reduces the need to perform additional model validation and development process. Nevertheless, supporting analysis methods - such as the heat utilisation efficiency indicator, used in parallel with other indicators - guide the designer in searching for new areas of optimisation and establishing the quality of the results.
  • Experimental and modelling studies on the impacts of fuels and lubricants on pre-ignition in spark-ignition engines
    (2025) Rönn, Kristian
    School of Engineering | Doctoral thesis (article-based) | Defence date: 2025-10-23
    This dissertation focuses on the issue of pre-ignition in spark-ignition engines. Pre-ignition occurs when the fuel-air mixture ignites prematurely before the spark timing, leading to excessive pressure rise and potentially uncontrolled pressure oscillations, also known as super-knock. Pre-ignition is a challenging problem because it occurs randomly, its initiation mechanisms are varied and complex, and it can lead to engine failure. Therefore, this work focuses on analyzing how pre-ignition manifests itself and what parameters related to fuels, lubricating oils, and the engines themselves can be adjusted to combat this phenomenon. The work includes a literature review, experimental research, and chemical simulations of fuels and lubricating oils. The applications of the research include turbocharged gasoline engines for automotive use and industrial scale hydrogen engines. It was concluded that pre-ignition can occur in gasoline engines through four different mechanisms: oil-fuel droplets, carbon deposits, autoignition of the bulk fuel-air mixture, and hot surface ignition. The literature review concluded that calcium-containing additives in lubricants, higher temperatures in the fuel distillation curve, and lower coolant temperatures promote the occurrence of pre-ignition. Gasoline engine tests focused on pre-ignition mechanisms that are independent of lubricants. The experiments showed that a high research octane number (RON) fuel does not guarantee good resistance to pre-ignition, because even high-RON fuels can be sensitive to pre-ignition caused by hot surfaces, as observed in tests with a fuel containing cyclopentane. However, both the pre-ignition tests and simulations identified a first-stage heat release phenomenon, where a small fraction of heat from the fuel-air mixture is released before the final ignition. This phenomenon consistently appeared under engine conditions where pre-ignition was observed, making it a reliable indicator of conditions prone to pre-ignition. It was found from tests with hydrogen-air mixtures and injection of various lubricants that an API Group V base oil combined with a calcium and magnesium containing additive provides better resistance to pre-ignition than a traditional API Group II base oil with a calcium additive. In another comparison, the base oil group was kept constant while the calcium content of the additive was changed. This change did not significantly affect the oil's susceptibility to pre-ignition, which contradicts findings from gasoline engine studies in literature. These results indicate a better potential for alternative base oil compositions to mitigate the pre-ignition problem in hydrogen engines compared to traditional mineral oil-based lubricants. Chemical simulations indicated that the doping of 2 mole percent of mineral base oil into hydrogen is sufficient to cause pre-ignition at a 13:1 compression ratio and 0.5 equivalence ratio. A corresponding amount of lubricant was also sufficient to cause pre-ignition at leaner equivalence ratios.
  • Ambivalence, dialogue, representation — Polyphonic approach to urban planning
    (2025) Harsia-Mikkola, Eveliina
    School of Engineering | Doctoral thesis (article-based) | Defence date: 2025-10-17
    Citizens’ experience of inclusion takes shape in the moments and situations where they interact with public authorities, such as urban planners. At the same time, a lack of democratic culture remains visible in urban planning. Although Finland introduced internationally ambitious participatory planning legislation as early as the 1990s, citizens’ actual opportunities to influence the planning of their living environments remain limited. Urban planning culture and institutional practices have been found to constrain these opportunities, often reducing participation to a formal exercise. This dissertation examines the disconnection between participatory perspectives and urban planning outcomes through the concept of polyphony. The aim is to deepen understanding of the meanings of polyphony in urban planning and to develop practices that can support more polyphonic planning. The research is situated within current legislative and digital reforms in Finnish urban planning, as well as the broader diversification of society, where increasingly plural communities and new forms of participation challenge traditional planning models. The empirical data consist of three case studies conducted in municipalities in the Helsinki metropolitan area: Nikkilä in Sipoo, Kontula in Helsinki, and Viiskorpi in Espoo. The cases address the role of participatory knowledge in valuing cultural environments, self-organized participation among entrepreneurs with immigrant backgrounds, and the construction of dialogical processes through hybrid participatory methods. Two of the cases employed an action research approach, in which the aim was not only to understand the phenomenon but also to influence it and develop practices in collaboration with participants. Methodologically, the research applies a multidisciplinary and multi-method approach, combining case studies, action research, and abductive reasoning. The research process is dialogical and iterative. The theoretical framework is grounded in planning theory, particularly in the tradition of communicative planning, which is further developed through the concept of polyphony, drawn from the literary theory of Mikhail Bakhtin. The dissertation presents a framework for polyphonic urban planning, which is used to explore the dialogical nature of participatory planning, its representations, and the agencies involved. The study argues that adopting and integrating polyphony into urban planning practices can strengthen a democratic planning culture. Polyphony is not only an abstract ideal or normative goal but also a practical tool for understanding and transforming the structures and agencies of urban planning.
  • Scaffolding novelty in product development: Enhancing idea generation and development
    (2025) Kirjavainen, Senni
    School of Engineering | Doctoral thesis (article-based) | Defence date: 2025-09-26
    The research presented in this dissertation investigates how novel idea generation and the further development of novel ideas can be effectively scaffolded through specific practices within product development, focusing on the phases of ideation, prototyping, and championing efforts. This work combines a framework from scaffolding literature, from the field of education and learning, with novel idea generation and advancement in the field of product development investigating ways to scaffold novelty. In doing so, the research aims to contribute to organizational innovation and competitiveness by systematically examining how developing and advancing novel ideas can be nurtured in product development. This dissertation consists of four articles, and utilizes a combination of qualitative and quantitative methodologies to explore scaffolding structures. First, qualitative document analysis was employed to deconstruct 127 documented idea generation methods, revealing underlying mechanisms that scaffold creative processes. Second, quantitative ideation experiments were conducted with 194 engineering students to assess the impact of these mechanisms on creativity metrics such as fluency, novelty, and originality. Third, thematic interviews with 43 product development professionals in a multinational industrial technology company provided deeper insights into the practical applications of prototyping and championing in the workplace. The results of the dissertation reveal the presence of multiple material and social scaffolds for scaffolding novel idea generation and further development of novel ideas. These are distributed across various levels of target and providing actors, from individual product developers to organizational management. These novelty scaffolds were found to be interconnected, indicating that their effective implementation requires a holistic approach that considers the complex dynamics and relationships within the product development process. Additionally, the research highlights several conditions that enhance the use of novelty scaffolds, such as organizational readiness, strategic resourcing decisions, skills development, and the leveraging of internal and external networks. The dissertation concludes by offering practical implications for product developers and managers, emphasizing the need for organizational commitment to fostering creativity through structured support mechanisms. The findings suggest that integrating purposeful prototyping early in the development process, encouraging iterative prototyping practices, and supporting championing efforts even after initial failure of an idea, are crucial for advancing novel ideas. Further, it underscores the importance of cultivating a culture that values experimentation and collaboration, providing the necessary resources and support to transform innovative ideas into successful products.
  • Machine vision analysis of vehicle interiors and surrounding road users in intelligent transportation
    (2025) Jayawickrama, Nilusha
    School of Engineering | Doctoral thesis (article-based) | Defence date: 2025-09-19
    Autonomous driving and shared mobility solutions have expanded rapidly, fuelled by continuous improvements in sensor technologies and deep learning. Intelligent vehicles require comprehensive awareness of (i) interior state, such as cleanliness, and (ii) exterior environment pertaining to neighbouring road user states. Achieving reliable insights into both spaces is essential for boosting safety and efficiency. Yet, a coordinated set of perception models that handle such specific interior and exterior entities under real-world variability has been lacking. This thesis aims to fill that gap with novel perception architecture implementations. For the interior task which relies on a stationary camera, a fine-tuned VGG16 network first classifies rear-seat scenes as clean, trash or valuables with 0.91 accuracy. The architecture is then extended with an EfficientDet-based detector which yielded prediction accuracies of 0.89-0.93 across broader variability (such as vehicle models and item types). The frameworks demonstrated effectiveness in automated cleanliness assessments within vehicle interiors. For the exterior task which relies on a dynamic (moving) camera, a scene-flow–guided framework fuses point-wise motion cues with tracking to exclusively identify nearby moving vehicles. It produced an F1-score of 0.95, underscoring the effectiveness of blending scene-flow vectors with tracking for improved motion prediction. Subsequent work expanded slot-attention based objectcentric learning (SA-OCL) to perception in autonomous driving for intelligent attention of road users, bridging the method's employability from generic video evaluations to a domain specific task with unique real-world complexities. Results proved the feasibility of the approach primarily to urban and city driving scenarios. Across four peer-reviewed articles, the dissertation answers seven research questions of visionbased techniques in an ego-vehicle pertaining (i) cabin cleanliness assessment with in-vehicle camera-based techniques (RQ 1-3; Publications I-II), (ii) scene-flow guided prediction of nearby moving vehicles (RQ 4-5; Publication III), and (iii) bridging SA-OCL to autonomous driving: priority-based attention of road users (RQ 6–7; Publication IV). Collectively, the thesis contributes: (i) the first end-to-end vision systems that enable in-vehicle scene-level and item-level predictions, (ii) a flow-vector-based motion prediction architecture that remains accurate under occlusions and multiple object interactions, and (iii) an SA-OCL based architecture for targeted attention of road users with minimal supervision, leveraging cues of proximity and motion. The outcomes of the thesis advance the field toward cleaner, safer, and efficient shared and autonomous mobility. Future work must refine the proposed methods for greater robustness across wider operating conditions for optimum scalability and real-world practical utility.
  • 3D-finishing of metal components manufactured by the powder bed fusion process
    (2025) Ullah, Rizwan
    School of Engineering | G5 Artikkeliväitöskirja | Defence date: 2025-09-19
    Laser-based Powder Bed Fusion (PBF-LB) is a promising additive manufacturing process for producing high-performance industrial components. However, achieving strict geometric product specifications (GPS) with dimensional tolerances of 10–50 μm remains a challenge, particularly for demanding applications such as injection molds, combustion engine pistons, and paper rolling machines. The absence of standardized machining allowances for PBF-LB components and the limited understanding of geometric deviations during post-processing further hinder its industrial adoption. To address these challenges, this study investigates machining allowances for 3Dfinishing of metal components manufactured via PBF-LB and examines the influence of process parameters on geometric deviations. A mixed-method approach is employed, integrating case studies, laboratory experiments, and finite element method (FEM) simulations. Full factorial testing is conducted for AlSi10Mg and maraging steel, examining anisotropy in mechanical thread strength and surface characteristics. Additionally, a previously unexplored thermo-mechanical FEM simulation model is developed and validated to predict cumulative geometric deviations during the PBF-LB process and after support removal process. The study reveals that AlSi10Mg exhibits higher warpage and anisotropy than maraging steel. However regardless of material, factors such as variations in energy input, staircase effect, and transition of melt pool stripe overlap to layer overlap impact the physical properties of end-products. Product orientation significantly impacts mechanical thread strength and surface properties. Machining allowances of at least 0.3 mm for linear features and 0.7 mm for curved surfaces are recommended. Additionally, the study finds that manufacturing setup misalignment contributes to geometric deviations alongside residual stresses. The validated FEM model predicts geometric deviations with an error margin of less than 30%, with further improvements possible through optimized mechanical contraction parameters. These findings inform comprehensive design guidelines that integrate machining allowances and predictive modeling for 3D finishing of PBF-LB metal components. A key limitation of this study is the need for a more robust method to accurately determine mechanical contractions induced during the PBF-LB process. Future research should explore material removal simulations, including side milling, to assess machining-induced deformations. Furthermore, incorporating setup misalignment and heat treatment simulations into FEM models could refine geometric deviation predictions for enhanced accuracy.
  • Evaluation of e-scooter usage in Helsinki Capital Region: System and user perspectives
    (2025) Dibaj, Samira
    School of Engineering | Doctoral thesis (article-based) | Defence date: 2025-09-19
    The rise of micromobility, particularly shared electric scooters (e-scooters), has introduced significant shifts in urban mobility, offering an alternative to motorized vehicles. The rapid deployment of shared e-scooters could positively and negatively impact the urban transport system. While research on e-scooter deployment is expanding, a structured analytical framework is lacking. This dissertation addresses this gap by employing a two-layered analytical framework to assess the implications of e-scooter deployment at both system and user levels. The data used in this dissertation is an online questionnaire survey in Helsinki Region containing sociode mographic characteristics of e-scooter users and non-users, trip purpose, reasons, their opinion on e-scooter deployment in Helsinki, perceptions towards e-scooters and their implications as well as mode substitution effect of e-scooters. In the first framework layer, to analyze the escooter deployment impacts and implications on the system-level mobility, a systematic literature review was conducted to understand the spatial and temporal distribution of e-scooter-related trips in the previous studies. Then, through a descriptive cross-tabular analysis, we investigated the escooter mode substitution effect on other transport modes. Furthermore, we analyzed the key factors that affect e-scooter mode substitution for six different modes in Helsinki using logit models. At the user-level layer, we tried to develop different e-scooter user and non-user profiles by deploying a systematic literature review and clustering methodology. The system-level analysis of e-scooter deployment indicates that e-scooters are predominantly used during midday and evening hours, particularly on weekends and in areas with recreational and educational land uses, city centers, and locations with well-developed walking and cycling infrastructure. In Helsinki, e-scooters were found to primarily substitute taxi/ride-hailing, bus/tram, and car trips. The user-level analysis of e-scooter users and non-users revealed that e-scooter users can be categorized based on their usage frequency and motivation, while non-users were classified into five distinct groups with varying degrees of negative perspective towards e-scooters, influenced by sociodemographic characteristics and motivations. Additionally, key factors affecting e-scooter mode substitution were identified, including usage frequency, private e-scooter ownership, trip purpose, reasons for use, age, income, and residential location. The findings contribute to a deeper understanding of e-scooter deployment’s system-level and user-level impacts, providing valuable insights for urban mobility planning and policy development.
  • Employees’ experiences in knowledge-based organisations: towards an age-friendly design and management of work environments for older workers
    (2025) Sandelin, Minna
    School of Engineering | Doctoral thesis (monograph) | Defence date: 2025-09-18
    The future workforce will increasingly be made up of ageing knowledge workers. This thesis explores the experiences of employees over 50 years of age and the viewpoints of management and experts in the context of knowledge-based work environments. The main theoretical field of the thesis is transdisciplinary workplace research, with a focus on experience research that combines physical, social, and technological workplace experiences. The overarching purpose is to develop further knowledge of employees’ experiences, and to explain which factors support older employees’ work and improve their work environment. This thesis adopted a qualitative research approach and a multiple case study design to understand how employees experience their work environments and which factors support their work and improve their physical, social and virtual work environments. Empirical data were collected by using a visual research method called probes including workshops and interviews with employees over 50 years of age (N = 77) as primary data; and, by conducting semi-structured interviews with management and experts (N = 29) as secondary data. The analysis of probe data was conducted in each case as a stand-alone entity, and the work environment data from each case were grouped into themes, allowing further interpretation through cross-case comparisons. The data from the semi-structured interviews were analysed as separate cases and interpreted based on cross-case issues of the physical, social and virtual work environments and the integrated workplace management. The results reveal many valuable employee experiences that affect people on the job and highlight several strategies used by management to design and manage work environments. It seems that employees are satisfied with the physical ergonomics in their respective offices but are concerned about ergonomics when working remotely. Cognitive ergonomics do not appear to be adequately supported with visual designs to aid cognitive abilities such as memory, learning, thinking and information processing. The findings highlight the need to pay closer attention to individual ergonomic needs. In addition, tacit knowledge should be more thoroughly considered to prevent the loss of critical organisational knowledge and experience as older employees exit the labour market. Surprisingly, work environments do not support social interaction. Further, a more flexible workplace, including flexible working hours and the possibility of remote work, would support the need for recovery from work and, in particular, the growing need to care for one’s own elderly parents. The results also show that by understanding work environment simply as a physical entity limits the potential of designing and managing age-friendly work environments. The findings highlight the need to manage across organisational boundaries through more integrated workplace management. This thesis contributes to the transdisciplinary workplace research and management with experience research approach. The phenomena the thesis focuses on is new and there has been little work done on studying the work environment for older workers. The thesis enhances understanding of the factors that support the work of older knowledge workers and improve their work environment. Such insights are valuable because future workforce will increasingly be made up of ageing knowledge workers in many of the developed countries. While this study confirms many factors related to the physical work environment, it also reveals that the concept of the workplace is not just a network of physical places but a broader experience–and that the physical work environment is only one aspect of the overall experience of a worker. This thesis gives voice to employees over 50 years of age and offers valuable knowledge for organisations regarding the design and management of age-friendly work environments; this knowledge will benefit the entire workforce, not just older workers. In addition, this research brings to the fore perspectives on employee experiences and the design and management of work environments in public discussions concerning the careers and extended working lives of older employees.
  • Risk management framework for maritime authorities: concepts, processes and maturity
    (2025) Laine, Valtteri
    School of Engineering | Doctoral thesis (article-based) | Defence date: 2025-09-12
    Maritime authorities have the administrative responsibility for managing the risk of accidental oil spills. This responsibility involves various regulatory-based activities aimed at maritime accident prevention, as well as pollution prevention and response. These risk management activities have also been supported with extensive research and professional efforts. However, in light of the identified needs of the maritime authorities, there is still a lack of risk management frameworks that consider the coexistence of multiple risk concepts in this field, the different risk management decision-making contexts of the authorities and the variations in their levels of risk management performance. The need for such frameworks is not only critical for practical purposes but also holds significant relevance from a theoretical perspective. To address the limitations of existing approaches, this thesis proposes a novel risk management framework tailored for the maritime authorities in the Baltic Sea region. Based on recent scientific research, industry standards and the practical needs of end-users, the framework adopts a risk-informed decision-making strategy and is built around three interrelated components. The purpose of its first component is to address the varying conceptualizations of risk within the maritime domain. Drawing on an interpretation of the Society for Risk Analysis's approach to risk, its objective is to establish a unified theoretical risk concept for this field, accommodate different perspectives on risk and provide a specific description of risk metrics. The purpose of the second component, in turn, is to account for the different decision-making contexts of the maritime authorities in managing the risk, as they range from short-term operational decisions to long-term strategic decisions. To achieve this, its objective is to provide systematic and flexible processes based on the ISO 31000 standard of the International Organization for Standardization, along with a set of tools and techniques to facilitate the associated risk assessment stages. Finally, the purpose of the third component is to enhance those risk management processes of the maritime authorities, including their essential prerequisites such as leadership, commitment and continuous improvement. By introducing a dedicated risk maturity model, known as the R-Mare matrix, it seeks to assist the authorities in assessing their current risk management performance and steering it toward higher levels of maturity. To evaluate the usefulness of this risk maturity model, its reliability is also tested with a maritime administration of the Baltic Sea region. Given the above, this thesis draws on state-of-the-art research to propose a new risk management framework that addresses the practical needs of maritime authorities in the Baltic Sea region and beyond. In this regard, it also supports efforts to apply recent developments in the theoretical risk field to create actionable risk management frameworks for practical use. Such efforts are strongly encouraged in research in this area.
  • Life-cycle analysis in timber construction - environmental impact and decision-making
    (2025) Niu, Yishu
    School of Engineering | Doctoral thesis (article-based) | Defence date: 2025-08-29
    The construction sector is recognized as a major player in climate change worldwide, requiring a shift towards renewable energy sources and sustainable material use. Timber, as a natural and renewable material, has been increasingly advocated as a replacement for carbon-intensive construction materials such as steel and concrete. The life cycle approaches such as life cycle assessment (LCA) are adopted, to address relevant benefits and obstacles from sustainability perspective. The main objective of this thesis is investigating the environmental performance of timber construction projects from life cycle perspective, to discuss relevant potential and challenges in climate change abatement. The research approach involves (i) an overview of existing LCA implementation on timber construction projects to find the research gap, and (ii) investigation on the selected overarching topics, including both upfront and post-use schemes of timber use in construction. The overview reveals large variation in the existing literature of LCA, as well as discovers common features. With respect to the large variation found in the overview, the research investigates the potential of using environmental product declarations (EPDs) to reduce the variability of LCA. However, the results show the obvious variation and inconsistency in EPDs for the same timber product. Similar finding is also observed for generic inventory datasets. Nonetheless, applying LCA to assist decision-making during the tendering process, with using EPDs, is explored. The results indicate that setting limits in the bidding document can effectively reduce the variability of LCA, meanwhile improve efficiency and comparability. Thus, consideration of environmental impact at design stage, as a decision rule to optimize design for sustainable construction is explored. The environmental impact is focused on climate change, depicted by CO2-eq. obtained from LCA results. The risk-informed optimization approach is utilized, which is cost-based. This approach is broadened covering environmental aspect, apart from structural safety and cost. A cost function reflects the above-mentioned aspects is developed. The influence of different end-of-life (post-use scheme) scenarios on the environmental impact of timber, is studied. The focus is wood cascading, addressing the relevant benefits and challenges in prolonging the life of timber and combating climate change. Three aspects are enclosed: technology, environment, and economy. The investigation reveals that policy is anticipated to be the driving force for the reuse of timber, in the Finnish context. Furthermore, it also indicates that, the equilibrium between reuse and energy recovery of timber should be accounted, as timber serves dual roles: material use for carbon storage and incineration for renewable energy, with each choice involving a trade-off.
  • Thermal Comfort of Older Adults in Hot Environments: Experimental and Numerical Investigation of Local Cooling Devices
    (2025) Chen, Minzhou
    School of Engineering | Doctoral thesis (article-based) | Defence date: 2025-08-29
    Older adults, as a vulnerable population, are more susceptible to heat-related mortality during heatwaves than younger individuals. In Finland, recent studies have highlighted this heightened risk. The room (without cooling systems) temperatures will exceed 32°C for about 3000 Kh. In addition, the declining income of the elderly also forces them to consider energy expenditure, thus facing more severe challenges. Some devices already on the market that can provide local cooling may help older people better maintain thermal comfort in hot weather. Compared with air conditioning, these devices are cheaper, more energy-efficient, easier to operate and more portable. However, there is little research on this topic. Thus, this thesis investigates the effectiveness of three types of local cooling devices in enhancing thermal comfort for older adults under various thermal environments. The research comprises four key components: 1) thermal responses of older adults in different thermal conditions; 2) changes in thermal responses after using three local cooling devices; 3) the physiological–psychological correlation of thermal sensation in older adults, and 4) modeling and validation of a novel personal cooling system. The results of the study showed that the neutral temperature of the elderly from northern Europe was 26 °C, their preferred temperature was 26.5 °C and the upper limit of the acceptable temperature was 28 °C. The lowest thermal acceptance rate was observed in environments with high temperature and relative humidity. At 28 °C (60%) and 29 °C (40%), the use of a table fan, evaporative cooling device, or air-cooled jacket could reduce the elderly's thermal sensation to neutral and make more than 80% of people accept the thermal condition. At 33 °C (40%) and 32 °C (50%), the use of an evaporative cooling device or aircooled jacket reduced thermal sensation significantly, but not to a neutral state. Although thermal acceptance rates increased after using all devices, they were less than 80%, except at 33 °C (40%), when evaporative cooling was used. Furthermore, all three devices performed better under conditions of lower relative humidity. Older adults' skin temperature in the head, limbs, and extremities has the strongest correlation with thermal sensation, and the thermal sensation in the head and torso exerts about 70% influence on overall thermal perception. Skin temperature and device usage had a complex causal relationship, and the cross-lagged effect between the two was the most significant at 5-minute intervals. The proposed novel local cooling system, which combined an evaporative cooling chair and a ventilated jacket can make older adults' thermal sensation reached to 0.5 in room at 33 °C. Meanwhile, it led to a notable 19 % reduction in electric energy consumption while maintaining similar elderly thermal sensation.
  • Quantifying Weld Undercut Severity on the Fatigue of Marine Structures
    (2025) Niraula, Abinab
    School of Engineering | Doctoral thesis (article-based) | Defence date: 2025-08-28
    Fusion welding is the most common metal-to-metal joining method in the manufacturing of metallic structures. The physical process in the fusion welding requires localized melting of the metallic components with the application of high energy, which then solidifies to merge and form the welded joint. The volatile nature of the welding process also introduces several imperfections in the welded structure, such as undercuts, which affect the mechanical performance and service life of the structure. Undercut imperfections have a significant detrimental effect on the fatigue life of welds, and they can lead to early failure of the structure, potentially risking public safety. Modern manufacturing methods can produce higher-quality welds without continuous undercuts; however, local undercuts are unavoidable due to the stochastic nature of the welding process. While the continuous undercuts have been widely investigated, the research on the effect of the local undercut geometry on fatigue has remained limited. A quantified approach is needed to assess the impact of different undercut geometries on fatigue. This dissertation aims to reveal the influence of the local undercut on the fatigue life and then to develop a characterization method for these undercuts. The research work systematically examines the effect of several individual geometric parameters of the undercut on the fatigue performance of the weld. Based on experimental observations, a parametric undercut geometry model is developed to describe the 3D geometry of the local undercut. Additionally, a geometric library of realistic undercuts is generated for a comprehensive numerical investigation. The research employs highresolution 3D geometry measurements, fatigue experiments together with statistical analyses and comprehensive numerical simulations. A strain-based fatigue life approach is utilized to investigate the relationships between the local undercut geometry and fatigue crack initiation life. The investigation covers varying local undercut geometries, materials, stress ranges, and weld geometries. The results of the thesis show that local undercuts exhibit a distinct 3D presence with complex geometries, which significantly influences the fatigue strength of welded joints. The shape of local undercuts was found to vary significantly, represented by the considerable fluctuation in the geometric parameters over short intervals of a few hundred micrometers. To characterize these complex geometries, a 3D Undercut Indicator has been proposed, demonstrating a correlation with the fatigue crack initiation life. However, this correlation is highly case-dependent, being sensitive to, e.g., load level and material properties. To address the sensitivity of this correlation, a novel, case-independent Undercut Severity Index (ΨUSI) has been developed. This index quantifies the fatigue-deteriorating effect of the local undercuts effectively, offering a unified approach to estimating fatigue crack initiation life by considering the influence of local 3D undercut geometry, applied stress range, material properties, and global weld geometry. The ΨUSI shows significant potential to promote a flaw-tolerant design philosophy and a fatigue-based weld quality control, offering a comprehensive toolkit for predicting the fatigue behavior of welded joints in marine, transportation, and civil industries.
  • Experimental studies on challenges in hydrogen heavy-duty and marine engines
    (2025) Yeganeh, Maryam
    School of Engineering | Doctoral thesis (article-based) | Defence date: 2025-08-08
    This dissertation addresses two critical challenges in developing hydrogen-powered heavy-duty and marine engines: (1) hydrogen jet dynamics and mixing and (2) hydrogen pre-ignition induced by engine lubricating oils. Experiments on hydrogen jet dynamics are conducted using high-speed z-type schlieren imaging in a constant- volume chamber under varying pressure ratios, needle lifts, nozzle geometries, injection angles, and injection durations. Results show that higher pressure ratios and increased needle lift enhance jet penetration and cross-sectional area, improving air-fuel mixing due to the greater mass of injected fuel and momentum. Nozzle geometry also significantly affects jet behavior. Single-hole nozzles produce faster penetration and multi-hole nozzles produce larger cross-sectional areas that improve air-fuel mixing. However, variations in the injection angle lead to jet-piston impingement near the periphery, resulting in uneven air-fuel mixtures, while injection duration has minimal impact on jet dynamics. These findings offer valuable insights for optimizing injection parameters to improve air-fuel mixing, especially in direct-injection hydrogen engines. The dissertation also investigates pre-ignition phenomenon in hydrogen engines induced by engine lubricating oils. Using a rapid compression expansion machine (RCEM) with optical access, this study evaluates lubricating oils from different API (American Petroleum Institute) categories. These lubricants, with varying levels of calcium and magnesium detergents, are studied under a range of air-to-fuel and compression ratios. The results reveal that API Group V base oils with lower calcium content exhibit greater resistance to pre-ignition, while the role of calcium in increasing reactivity in API Group II oils is minimal. These findings highlight the critical role of engine lubricating oil composition in influencing flame propagation, pre-ignition limits, and overall combustion dynamics. By integrating experimental and computational approaches, this dissertation provides practical solutions for optimizing jet dynamics and mixing, as well as for selecting lubricating oils for hydrogen engines. These advancements support the adoption of hydrogen as a zero-carbon fuel for heavy-duty and marine applications.
  • Team teaching in multidisciplinary technology education: Instructional process in design-based technology projects
    (2025) Aarnio, Hanna
    School of Engineering | Doctoral thesis (article-based) | Defence date: 2025-06-30
    Technological society increasingly requires crossing disciplinary boundaries. Educational institutions, ranging from primary to higher education, seek to educate future experts who can combine technological expertise and perspectives of different disciplines in the collaborative creation of technological innovations. To support pupils’ multidisciplinary learning of technology, teachers need to identify and leverage the yet less explored forms of collaboration in technology education. This dissertation explores how team teaching enables multidisciplinary, design-based technology education at primary, secondary, and upper-secondary levels. Team teaching is studied from two perspectives: 1) the instructional process relying on team teaching in design-based technology projects and 2) social support between colleagues as an enabler for team teaching in multidisciplinary technology education. The thesis consists of three part studies that applied interviews and longitudinal classroom video data. The participants were altogether 21 teachers from 11 teaching teams and six pupils. The 11 teaching teams were interviewed in 2019–2021. At the time of the interviews, teachers participated in a continuous education program pilot organized in collaboration between technical higher education and teacher education institutions. The video data were collected in 2021 from a board game design project in a primary school. The participants were two experienced crafts teachers and six pupils who worked in two small groups. The analysis for the interviews followed the principles of qualitative content analysis. The video analysis adopted the three-level approach, including macro, meso, and micro level, and the outcomes were visualized by the Making-Process-Rug method. The findings of the three part studies result in an overview of support for collaborative teaching and learning in technology education. The key contribution to the existing body of knowledge in multidisciplinary technology education is an empirical model of the collaborative instructional process. The model shows that in the planning of the projects, the teaching teams’ focus was on support for materials and technologies, while in the implementation, the emphasis was on scaffolding and epistemological support for learning. In assessment, teachers led diagnostic assessment activities. Formative and summative assessments were prevalent among pupils and connected to verbal and embodied design activities. Moreover, in the planning phase of the instructional process, social support between colleagues in a team served as an instrumental resource for technology education, enabling versatility for implementing the projects. Three major conclusions are drawn based on the findings. 1) Teacher collaboration provides diverse forms of support for pupils’ learning of technology during the instructional process of design-based technology projects. 2) Formative assessment between pupils is a valuable yet overlooked resource in supporting design activities in technology projects. 3) Teachers’ collegial instrumental support increases the versatility of ways to implement technology education, but utilizing its full potential requires pedagogical leadership. The recognized forms of collaboration in multidisciplinary technology education can be applied by teachers to strengthen pupils’ technology study paths across the levels of education, thereby enriching the pool of future technology professionals.
  • The Finnish Civil Defence shelter system - Evolution of the regulation and technical specification 1954-2011
    (2025) Kyrenius, Pekka
    School of Engineering | Doctoral thesis (monograph) | Defence date: 2025-06-17
    Finland has approximately 50 500 Civil Defence (CD) shelters with 4.8 million shelter places for a population of 5.6 million residents. CD shelter building is mandatory by legislation, and Acts, Decrees, and Technical regulations regulate it. This study examines the evolution of the Finnish CD shelter regulation from 1954 to 2011. The development of shelter regulation is affected by military threat assessments and the commitment to protect the population cost-effectively from the effects of weapons during a potential wartime. This includes the objective of providing a relatively safe and healthy, but uncomfortable environment for sheltering periods. The method draws on comparative historical analysis, comparing the Finnish shelter design baseline with the Swedish, Swiss, US, and German baselines, incorporating respective analyses of changes along the study timeline. Change analysis includes iteration and backtracking along the timeline. The ventilation and thermal control baselines are evaluated using established technical methods. This study defines the Finnish CD shelter systems as a socio-technical system that includes diverse actors, interest groups, organisations, regulations, and the complete nationwide materiality of the CD shelters. The Large Technical Systems (LTS) theory is used as an analytical framework. With the help of socio-technical analysis, this study focuses on two research questions: What significant changes are evident in the shelter regulation timeline, and what is the root cause or impact of these changes? Are there any significant gaps in the Finnish shelter studies or regulations, and what is the effect of these gaps? This study's contribution to LTS research is proving that a mature system exhibits different change dynamics as regulation justification is contested. On the other hand, a mature system may detach from regulation or the prevailing context. The evolution of regulation is studied with three case studies focusing on weapons effects mitigation, shelter occupant life support, and combining both focal points. This study establishes how the Finnish CD shelter system reached its technical style. The noteworthy Finnish attribute, in reflection of most other CD shelter-building nations, is the cost-effective funding model. The building owner bears the cost of the shelter construction, not the taxpayers. The Finnish full-scale shelter habitation tests with occupants, together with the analysis of this study, have proven that the Finnish shelter design baseline is sound. The shelter’s thermal control and air conditioning during the hot and humid summer months advocate further studies. The Finnish CD shelter system is a success story. With an average of three years spent on military defence, a shelter system has been built cost-effectively in six decades to protect the whole population. The estimated aggregated cost of the Finnish CD shelter system is approximately 4.4 milliard euros.
  • Scalable and Robust Machine Learning Solutions for Adaptive Building Operations
    (2025) Stjelja, Davor
    School of Engineering | Doctoral thesis (article-based) | Defence date: 2025-06-13
    Achieving a more efficient, resilient, and low-emission energy system is crucial for long-term sustainability. One of the primary paths for this transformation is the digitalization of energy systems and buildings, which has led to a surge in operational data. Machine learning (ML) methods can leverage this data to provide deeper insights, automate complex processes, and reveal new opportunities in building management. Specifically, ML methods are applied in this dissertation to predict energy consumption, estimate building- and room-level occupancy, and detect changes in consumption patterns. This dissertation addresses two main challenges in applying ML to these building operation applications: scalability and robustness. The first goal is scalability, ensuring that the developed ML solutions can be applied across various buildings, each with its own unique characteristics and operational dynamics. Achieving scalability means minimizing the need for extensive ground truth data collection, which can be costly and time-consuming. By reducing reliance on ground truth data, the dissertation tries to make these solutions more practical and transferable across different building environments. The second key goal is model robustness, the ability of machine learning systems to consistently perform well, even when faced with changing environments and varying operational conditions. This characteristic is fundamental for systems deployed in dynamic building environments. Furthermore, deviations from robustness can reveal underlying shifts in system behavior. This objective focuses not only on developing models that adapt to such changes, but also on utilizing robustness failures as signals to inform and enhance decision making in real-world scenarios. To meet these goals, the study utilizes building data on energy consumption, sub-metering, indoor temperature, and CO2 levels. Scalability was enhanced through several strategies: unsupervised learning (clustering) was applied to infer occupancy from sub-metered consumption withoutground truth data, while transfer learning enabled room-level occupancy prediction using only a few days of labeled CO2 and temperature data, significantly reducing the dependence on extensive ground truth. Additionally, domain knowledge was incorporated into a probabilistic energy consumption model, improving its ability to detect meaningful shifts and potential anomalies without explicit labels. Robustness was improved by employing transfer learning combined with sliding normalization, enabling model to adapt effectively to operational changes, such as adjustments in ventilation settings. Additionally, deviations in model performance were leveraged as signals to identify anomalies, utilizing domain-informed probabilistic modeling. Tests carried out with low to moderate noise in the training data confirmed that the proposed method maintained reliable performance. Overall, this research demonstrates the feasibility of scalable and robust ML solutions that capitalize on readily available building operations data. Accurate occupancy estimates, energy consumption predictions, and anomaly detection can be achieved with minimal reliance on extensive ground truth, paving the way for more data-driven, efficient, and robust building management. The work shows that it is possible to balance scalability and robustness by designing solutions that generalize across diverse buildings while remaining responsive to changing operational conditions.