Aluminum castings have been widely used in the automotive industry to reduce the vehicle's weight. However, the existence of casting defects significantly limits its application. The most common and detrimental defects in aluminum castings are porosity and oxides. The formation of the pores comes from the solute hydrogen and volumetric reduction during the solidification process, resulting in the gas and shrinkage pores, respectively. The oxides can be introduced by either the operation during the process or the originally existing oxides in the melt. To reduce these defects, optimizing the casting process and controlling the melt quality is essential.

In this work, the Semisolid Metal (SSM) process was used as it can significantly reduce the formation of shrinkage pores. The main focuses were on the influence of stirring intensity on the formation of casting defects and, thus, the resultant mechanical properties. In addition, to control the original melt quality, particularly the amount of oxides, efforts were made to develop proper methods for the melt quality assessment.

The results show that the slurry-making process, mainly through stirring intensity, can affect the casting defects significantly. On the one hand, the increasing stirring intensity can distribute the primary Al particles more homogeneously, reducing the pores in terms of size and number by increasing the permeability during the solidification process. On the other hand, the increasing stirring intensity can affect the size of oxides differently, depending on the composition, for instance, the Mg content.

For the alloys with sufficient Mg, the oxides would be MgAl2O4, which are small films with numerous cracks. Under intensive stirring, the oxides can break down into small oxide particles. As a result, intensive stirring can improve ductility by reducing the formation of big pores. However, the oxides would mainly be Al2O3 films for alloys with low Mg content. In this case, the current stirring intensity is insufficient to break the oxide films. Instead, the increased stirring has introduced more oxide films into the melt. Consequently, in the casting with intensive stirring, the increasing oxide films dominated the ductility rather than the reduced porosity.

The SSM castings exhibit better bending fatigue properties than the casting using the traditional high-pressure die casting (HPDC) process. This improvement is mainly due to the formation of the harder surface liquid segregation (SLS) layer on the SSM casting surface. Furthermore, compared with the standard SSM process, the castings using intensive stirring (hereinafter referred to as the modified SSM process) show similar but more reliable fatigue properties. This reliable fatigue property can be attributed to eliminating the big internal pores through intensive stirring, which results in local stress concentration and significantly reduces fatigue performance. Besides, due to the gradient stress distribution in the bending loading, the surface defects play a significant role in the fatigue properties. With the increase of the specimens’ thickness, the failure mechanisms changed.

The shrinkage pores in the reduced pressure test (RPT) test play a significant role in the accuracy of melt quality assessment. A good correlation between the bifilm index (BI)/ density index (DI) and hydrogen content is observed for the RPT samples without significant shrinkage pores. In addition, the correlation between the BI and elongation is also strongly affected by the clusters of shrinkage pores due to the conflict between the definition of the BI and the influence of clusters of shrinkage pores on the ductility. Based on this, we proposed an optimized BI where the clusters of shrinkage pores were treated as single pores, increasing the reliability of the correlation between the BI and elongation. 

This thesis focuses on promoting patient safety in older persons using medications. Given that medications can unintentionally harm patients, the World Health Organisation emphasises “Medication without harm” as a global patient safety challenge. Older persons are more likely to experience harm, and harm tends to occur when prescribing or monitoring medications. Co-production of healthcare with patients may reduce the risk of adverse events and can serve as a resource to promote safety in healthcare. Accordingly, this thesis aims to increase knowledge of how older persons and healthcare professionals can co-produce a solution for improved medication evaluation and thereby promote patient safety.

Co-design is an approach that emphasises patient involvement in improvements of healthcare services. Therefore, the thesis was guided by the four phases of the Double Diamond framework for co-design. In the Discover phase experiences of medication evaluations were collected. Older persons were interviewed and data were analysed using qualitative inductive content analysis (Paper I). Critical Incident Technique was used to collect and analyse data from interviews with healthcare professionals in primary care (Paper II). In the Define and Develop phases, a case study design was used to explore older persons’, nurses’ and physicians’ design choices for a medication and their experiences of a remote co-design approach. Collected data were analysed using descriptive statistics along with directed content analysis (Paper III) and thematic analysis (Paper IV). In the Deliver phase, the feasibility of applying a medication plan in primary care, as well as the study methods used were examined. Data were analysed using descriptive statistics and inductive content analysis (Paper V).

The findings showed that older persons reported having a responsibility to engage in their medication evaluations, even if some felt unable to do so or considered themselves unconcerned. Continuity of care and participation facilitate evaluations, but a comprehensive medication evaluation was lacking (Paper I). Healthcare professionals experienced that medication evaluations for older persons were influenced by working conditions and working in partnership. Actions taken to manage medication evaluations were carried out through working with a plan and collaborative problem-solving (Paper II). A medication plan, linked to the medication list, had to provide an added everyday value related to safety, effort and engagement, and support communication, continuity and interaction. Important functional requirements were to provide instant access, automation and attention, and content requirements were detailed information about the medication treatment (Paper III). Remote co-design can complement or substitute for face-to-face co-design sessions. The approach allowed an accessible environment, and sharing everyday life experiences created learning and awareness of possible risks and strategies that could promote patient safety (Paper IV). The feasibility of applying a medication plan, assessed as usability, varied and the participants’ experiences of usability concerned a de-prioritised medication plan, functionalities, individualisation and resources. The participants’ perceptions of patient safety addressed awareness and information, challenges beyond the medication plan and patient involvement (Paper V).

Healthcare services could promote patient safety by involving older persons in medication evaluations and in co-designing patient safety solutions. However, implementing a medication plan in clinical practice is complex and requires continuous co-produced improvements at different levels within the healthcare system.

Industrialized house building (IHB) needs to improve efficiency and reduce costs to be able to meet the demand of future building requirements. The discrete manufacturing industry has already begun its transformation with concepts such as Industry 4.0 and smart manufacturing. The focus of these concepts is on systems integration and utilizing different design principles and technologies to improve productivity and competitiveness. So far, smart manufacturing has not been discussed to any significant extent in the context of the IHB industry, even though it might be a way for this industry to improve efficiency and reduce costs.

The purpose of this thesis is to support smart manufacturing realization in the IHB industry through improved systems integration capability. The objective is to develop a support system that helps to mitigate current product data management challenges in the IHB industry. The research presented here followed the design research methodology (DRM), and three research studies were conducted within the IHB industry, specifically within the context of the wooden single-family house industry.

The findings emphasized that integrated information and communication technology (ICT) tools are a precondition for realizing smart manufacturing in IHB. A pilot product lifecycle management (PLM) system was developed as a support system to mitigate the challenges encountered in product data management. The support system was developed based on five requirements (integrate product data between ICT tools, extract product data to purchasing items, adaptable to the product realization process, organize product data in different views, and trace revisions on building information modeling (BIM) projects and BIM families) and six functions (extract and interpret product data from the BIM project, property mapping, revision management, different types of document structures, from document structure to item structure, and complete bill of materials). The pilot PLM system was perceived to have value and could, for the case company, mitigate their challenges with product data management and facilitate improved systems integration. Furthermore, 18 enablers for PLM implementation in IHB were identified and categorized into implementing organization, PLM implementation process, PLM system selection and development, and development of product data for PLM.

The use of Artificial Intelligence (AI) has transformed fields like disease diagnosis and defence. Utilising sophisticated Machine Learning (ML) models, AI predicts future events based on historical data, introducing complexity that challenges understanding and decision-making. Previous research emphasizes users’ difficulty discerning when to trust predictions due to model complexity, underscoring addressing model complexity and providing transparent explanations as pivotal for facilitating high-quality decisions.

Many ML models offer probability estimates for predictions, commonly used in methods providing explanations to guide users on prediction confidence. However, these probabilities often do not accurately reflect the actual distribution in the data, leading to potential user misinterpretation of prediction trustworthiness. Additionally, most explanation methods fail to convey whether the model’s probability is linked to any uncertainty, further diminishing the reliability of the explanations.

Evaluating the quality of explanations for decision support is challenging, and although highlighted as essential in research, there are no benchmark criteria for comparative evaluations.

This thesis introduces an innovative explanation method that generates reliable explanations, incorporating uncertainty information supporting users in determining when to trust the model’s predictions. The thesis also outlines strategies for evaluating explanation quality and facilitating comparative evaluations. Through empirical evaluations and user studies, the thesis provides practical insights to support decision-making utilising complex ML models.

Global warming is driving industry to manufacture lighter components to reduce carbon dioxide (CO₂) emissions. Promising candidates for achieving this are aluminium-silicon (Al-Si) cast alloys, which offer a high weight-to-strength ratio, excellent corrosion resistance, and good castability. However, understanding variations in the mechanical properties of these alloys is crucial to producing high-performance parts for critical applications. Defects and oxides are the primary reasons cast components in fatigue applications are rejected, as they negatively impact mechanical properties.

A comprehensive understanding of the correlation between fatigue performance and parameters such as the α-aluminium matrix, Al-Si eutectic, surface roughness, porosities, hydrogen content, oxides, and intermetallic phases in Al-Si castings has not been reached.

The research presented in this thesis used state-of-the-art experimental techniques to investigate the mechanical properties and crack-initiation and propagation behaviour of Al-Si-Mg cast alloy under cyclic loading. In-situ cyclic testing was conducted using scanning electron microscopy (SEM) combined with electron back-scattered diffraction (EBSD), digital image correlation (DIC), and focused ion beam (FIB) milling. These techniques enabled a comprehensive study of parameters affecting fatigue performance, including hydrogen content, surface roughness, oxides, and intermetallic phases. More specifically, we investigated the effect of melt quality, copper (Cu) content, oxide bifilms, surface quality, and porosity.

The increased Cu concentration in heat-treated Al-Si alloys increased the amount of intermetallic phases, which affected the cracking behaviour. Furthermore, oxide bifilms were detected at crack-initiation sites, even in regions far away from the highly strained areas. Si- and Iron (Fe)-rich intermetallics were observed to have precipitated on these bifilms. Due to their very small size, these oxides are generally not detected by non-destructive inspections, but affect mechanical properties because they appear to open at relatively low tensile stresses. Finally, Al-Si alloy casting skins showed an interesting effect in terms of improving fatigue performance, highlighting the negative effect of surface polishing for such alloys.