Recent advancements in artificial intelligence (AI), particularly in deep learning, have significantly enhanced AI capabilities but have also led to more complex and less interpretable algorithms. This research addresses the challenge of Explainable AI (XAI) by focusing on enhancing the interpretability of AI decisions through the use of Explainable User Interfaces (XUI). The study identifies two primary knowledge gaps: the predominance of XAI research targeting technically skilled users, neglecting the end-user who often lacks technical expertise, and the insufficient exploration of user-centric design principles in real-world XUI applications.

The research adopts the Design Science Research Method (DSRM) to develop an XUI tailored for the FOKUS project, which uses Electrocardiogram (ECG) data to detect myocardial infarctions. The study emphasises the strategic application of interactive design principles such as complementary naturalness, flexibility in explanation methods, and responsiveness through progressive disclosure to improve the system’s interpretability. Notably, sensitivity to context and mind, though not initially implemented, emerged as a critical design principle from the analysis and was subsequently positioned at the pinnacle of a restructured pyramid model of design principles.

Key findings highlight the effectiveness of the selected design principles in enhancing interpretability and underscore the importance of involving stakeholders early in the development process to align the XAI and XUI with end-user needs. The research proposes a structured design approach framework for XUI, involving sequential phases from pre-XAI to XUI design, to systematically integrate user feedback and improve the design iteratively. The proposed framework restructured pyramid model of the design principles aim to guide future developments in XAI and XUI, enhancing their practical application and effectiveness in various contexts.

This master's thesis addresses a common problem of the adoption of IT automation; the nonoptimal use of the technology due to ineffective managerial communication. It searches for a solution by using the Design Science Research methodology to create a design artifact in the form of a managerial communication framework. This framework provides values to all stakeholders in the IT automation adoption process; the managerial level, the employees, and the technology provider. It improves managerial communication during the adoption process and identifies actionable guidelines for managers.

Narrowing down the scope of the thesis to the adoption of Robotic Process Automation (RPA) within the recruitment industry allows the research to be manageable within the constraints of a master's thesis. A literature review is used to investigate the knowledge base of the topic of the thesis. Additionally, semi-structured interviews with managers and employees are used as a means of collecting data from the environment. Combining these inputs, this research designs a model that explains the role of managerial communication in the effective adoption of IT automation. The model consists of four main components: managerial communication, motivation & vision, impact awareness, and the effective adoption of IT automation. It explains the constructs that were found relevant in this process and how they relate to each other. Several root constructs are used as a basis to create actionable guidelines to facilitate an effective adoption process. Together with the explanatory model, these guidelines form the managerial communication framework. The main goal of the framework is to bridge the gap between the manager's and the employee's views on the nature of the change. This study finds that managers often view the change as a process optimization, whereas employees view it as an organizational change. The managerial communication framework allows to identify these differences and act upon them.

This study not only designs the communication framework, but also evaluates its defined objectives with several experts. The results of these evaluation interviews demonstrate the validity and usability of the research conducted and the design artifact. Future research can extend the framework to other domains & technologies and tailor it to specific problem cases. Furthermore, the framework can be researched and evaluated in practice by conducting a case study.

The metaverse is one of the most disruptive technologies to evolve from the digital transformation. While the potential use cases of creating an immersive virtual world are numerous, the vision of an industrial metaverse is only recently emerging as a concept from the technology. In the automotive sector, manufacturers are starting to use simulation, digital twin technology and Building Information Modelling (BIM) to build virtual factories in an industrial metaverse. The benefits of this innovation are believed to significantly boost production flexibility and efficiency, which is why manufacturers set up data-driven digital platforms to enable an industrial metaverse that interconnects multiple actors. How-ever, technical barriers still hamper the implementation of such platforms whose dependence on flawless data grows with the number of use cases for an industrial metaverse. Accordingly, quality insufficiencies of spatial data and the absence of automatic quality assessments to identify these insufficiencies are one of the most decisive barriers to a widespread adoption of industrial metaverse applications. This thesis examines this problem and investigates how data quality insufficiencies in an industrial metaverse en-vironment can be identified and overcome at the example of an automotive manufacturer that uses the Nvidia Omniverse digital platform to create virtual factory models. A design science approach is pursued to create an extension to the Omniverse software that identifies the most critical data quality insufficien-cies, derives key performance indicators (KPIs) and proposes preventive measures to induce a sustained data quality improvement. Thereby, this thesis lays the groundwork for future research emerging around the concept of an industrial metaverse and the remaining obstacles of digital platforms to enable its applications. The pursued DSRM approach to overcome such barriers is capable to serve as guidance for future research projects that pave the way for a gradual enablement of further industrial metaverse use cases in other industries.

Yearly, 20.000 fatalities are recorded in road accidents on European roads alone. This situation is far more severe (1.3 million humans) when the global scope of road accidents is considered. Thus, Policymakers are initiating policy programs to overcome this challenge by enhancing safety and maximizing the associated positive economic impact. The EU committed to minimizing fatalities by 50% by 2030 and to be net zero long-term by 2050.

To achieve this, vehicular communication networks (VCN) can play a crucial role. It is anticipated that the connectivity and communication of vehicles with their environment contribute to these goals by enabling safety functionalities or improving the traffic flow. Hereby, it is crucial to understand that this communication requires an infrastructure that assists the vehicles and ensures data transmission and operation. However, due to the European size and heterogeneity of the member states, realizing a vehicular communication network on a large scale requires increased coordination and collaboration efforts. Currently, no vehicular communication network has been introduced working EU-wide. Hence, this resembles a potential for future road improvement. To realize this, decision-makers must understand the complex properties and system interrelationships of such a large-scale infrastructure project so that a common VCN can be designed to ensure interoperability and robust functionalities across the EU. This work contributes to the aforementioned challenge by addressing the following research question:

What are fundamental socio-technical factors to consider in a future European vehicular communication network design?

A mixed approach that combines Peffer´s design science research framework and system engineering methodology is used to synthesize the contributions. By this, a vehicular communication network mission, stakeholder, and system analysis are presented in this work. Further novel and scientific sound requirements and stakeholder insights are synthesized by systematically reviewing 57 articles and interviewing 15 experts from the institutional, scientific, and industry domains. In this context, a stakeholder classification for vehicular communication networks, a 4-layer stakeholder complexity model, and a system requirement structure from a system perspective are proposed to contribute to the VCN understanding and future design attempts. Further, the reflection of the socio-technical interrelations between the technical and social VCN subsystems are subjects of this work.

The conclusion is that the reflection on socio-technical system properties plays a critical role in vehicular communication network design. Further, a future vehicular communication network consists of a magnitude of stakeholders with high interest and power; thus, designers must understand the characteristic of their co-evolutionary cooperative development. Hence, a multidisciplinary understanding and approach are critical for designers. Furthermore, the geographical segregation of designing and decision-making in a future vehicular communication network is identified, and certain goals/issues should be addressed in the respective layer. Another major conclusion is that the VCN discussion is determined more by social and socio-technical conditions, such as stakeholder cooperation/coordination and interoperability, than by technical feasibility.
These findings result in three main research contributions, which are summarized as follows:

- A 4-layer stakeholder complexity model contributing to the understanding of VCN development is contributed.
- An approach of integrating the socio-technical system perspective on the complex, large-scale infrastructure VCN project is contributed. This focuses on the processes and requirements between social and technical subsystems, addressing the integration of heterogeneous stakeholder interests.
- A comparison is made between scientific focus, stakeholder needs and objectives, and expert insights, highlighting the mismatch and alignment of requirements. This contributes valuable insights for adjustment and further research in the VCN community.

Further methodological contributions can be concluded:
- The design science research approach is aligned with systems engineering iso standards and methodology.
- The design science research framework is used as an approach to address the complexity of VCN systems. This novel perspective helps VCN stakeholders design solutions for their field problems.

The results have implications for designing stakeholders in a VCN. Based on the analysis, it is recommended that policymakers identify and extend common objectives with the industry to establish public-private business cases. Further, vehicle manufacturers should participate and embrace the transition to a vehicular communication network by cooperating strongly with stakeholders. In addition, implementing the relevant technologies to communicate with the heterogeneous infrastructure is suggested to shape future infrastructure connectivity development. Lastly, the lack of socio-technical reflection in scientific literature is identified. Hence, scholars should elaborate on the interactions and interrelationships between the social and technical subsystems in future work.

Explainable AI (XAI) systems are rapidly gaining significance. While frameworks for XAI interpretability for experts abound, metrics for laypeople’s comprehension are absent. This study addresses this gap by investigating interpretability factors from both developer and layperson perspectives. The core research question is: "How can XAI developers assess to what extent XAI is interpretable for laypeople?".

Applying the Design Science Research Methodology, findings from multiple literature reviews are combined to construct a preliminary XAI interpretability framework for laypeople, featuring crucial factors and their relationships, as well as associated principles. The proposed framework underwent validation through semi-structured interviews with 12 XAI experts, informing revisions and refinement of our key principles. Subsequent layperson surveys, considering a specific use case, offered insights into preferences about interpretability factors, informing further refinement.

The final theoretical framework highlights pivotal factors including simplicity, transparency, comprehensiveness, complexity, clarity, generalizability, trustworthiness, explanation fidelity, model fidelity, intentionality, relevance, affordance, coherence with prior beliefs, and actionability. Surrounding the framework are key principles emphasizing trustworthiness, relevance, simplicity, clarity, coherence, intentionality, actionability, fidelity, contextualization, and ethical considerations, serving as actionable guidelines for XAI developers and researchers.

The implications of the study are profound, offering valuable insights for advancing XAI research and system design. The refined framework and principles act as a foundation for both novice and experienced XAI developers, fostering interdisciplinary research among AI, human-computer interaction, psychology, and philosophy experts. These findings can drive the responsible adoption of AI systems across sectors like healthcare, finance, and transportation, while informing policies and regulations governing AI technologies. Our study promotes responsible AI practices, enhancing user trust and understanding, while facilitating the creation of more effective guidelines and standards.

Modern manufacturing operations rely on accurate, real time data for efficient operation and control of the production process. Many small and medium sized enterprises (SMEs) still operate with legacy assets, limiting their access to a clear data picture. This thesis research and its design solutions help SMEs gain real time insights into their manufacturing process, resulting in production efficiency gains and helping identify process improvements. These factors help manufacturing SMEs stay competitive in an ever-increasing competitive global business climate.

The primary objective was to identify the requirements for enabling real-time insights into SME manufacturing operations. Findings suggest that in-house infrastructure can facilitate the required transparency, provided it is lightweight and minimalistic. The essential factor is to transform machine-generated data into insightful metrics for relevant stakeholders. A significant aspect to consider is the human element— the adoption and commitment of employees towards the technology.

Upon analyzing the benefits, the research underscores that real-time insights lead to enhanced production efficiency. By offering an objective view into production data, SMEs can quickly identify and rectify bottlenecks. Moreover, such transparent insights improve communication within teams, empowering employees by showing the direct impact of their performance on overall production, thus boosting their morale and productivity.

The study also evaluates the key features needed to make these real-time insights actionable. Contrary to initial assumptions that a single KPI would suffice, it became evident that a more comprehensive view—incorporating metrics such as Overall Equipment Effectiveness (OEE)—was necessary. For SMEs with diverse product portfolios, additional parameters like average batch size, cycle time, and machine power-on time provide valuable context.

Lastly, the implemented design framework showcases the positive impacts of real-time insights on production. Access to objective metrics aids in performance evaluation, problem identification, and the formulation of improvement strategies. An essential feature highlighted is the predictive capability that guides resource planning, thus minimizing errors.

This research emphasizes that manufacturing SMEs with legacy assets can achieve real-time insights into their processes through cost-effective, lightweight solutions. The proposed methods and guidelines, as discussed in detail in the design section, are foundational for SMEs aiming to modernize their operations and bridge the knowledge gap in implementing real-time monitoring. By offering universal access to transparent production metrics, SMEs can optimize their processes, ensuring better efficiency and communication at all levels of operation.

The fourth industrial revolution is upon us, and entire industries are seeking to reap the benefits resulting from the use of the technologies that the revolution brings. Waste management is one of those industries. The rapid growth of the human population causes more consumer goods to be produced every day. This underlines the importance of effective and efficient waste management.

The Internet of Things (IoT) has proven to be a useful tool to increase the efficiency of waste collection and while many waste management organizations are beginning to adopt these solutions, many are struggling to fully adopt and implement the technology. It is still unclear, due to lack of research, what factors are hindering the adoption and implementation of IoT technology in these organizations and how the process can be improved. The research scope of this study was three-sided: Industry specific, organizational, and technological. The focus was set on the waste management industry, and the intra-organizational barriers that hindered the adoption and implementation process of IoT-powered fullness sensors. The core problems at hand were identifying what the intra-organizational IoT adoption barriers are, what are the most influential barriers, what mitigation strategies can be employed to mitigate these barriers and how it can all be illustrated within an IoT adoption and implementation process framework.

The main research question formulated to answer these problems is: How can the adoption and implementation process of IoT-powered fullness sensors in waste management be improved? Four sub-research questions were formulated that held partial information which were needed to answer the main question.

The overarching structure of this research project follows the Design Science Research Methodology (DSRM). DSRM provides a commonly accepted approach which involves a rigorous six step activity process for creating and evaluating an IT artifact intended to solve organizational problems. This methodology was altered to fit this particular study, and thus followed five of the six steps. An analysis of literature was performed to identify the general intra-organizational innovation adoption barriers. An exploratory case study was conducted within a large waste management company in Iceland which recently decided to install IoT-powered container fullness sensors to increase the efficiency of their processes. The case study revealed, through interviews, which of the identified barriers had the most significant effect on their adoption and implementation process. Expert interviews and desk research were used to formulate strategies that organizations could employ to mitigate the most prominent barriers identified. After all the interviews had been transcribed, coded, and categorized, data triangulation was used where data from multiple different interviewees was compared and analyzed. From the results of these research efforts, a framework explaining the IoT adoption and implementation process for waste management was then designed and developed. Expert interviews were again conducted to evaluate the framework and confirm the framework’s theoretical validity, application and expected performance in terms of its set goals and objectives.

The results of this research are the identified general intra-organizational innovation adoption factors, the most influential factors affecting IoT adoption and implementation within waste management along with their proposed mitigation strategies as well as a designed IoT adoption and implementation process framework in which these strategies are incorporated. The general intra-organizational innovation adoption factors identified are: Leaders’ attitude towards change, Centralization, Complexity, Formalization, Interconnectedness, Organizational slack, Size, Culture, Degree of risk-taking, End user behavior, Strategic objectives and Uncertainty of business benefits. The following are the three most influential barriers to IoT adoption and implementation within waste management and their proposed mitigation strategies: The first barrier is Uncertainty of business benefits and its proposed mitigation strategies are: Gaining a Proof of Value (PoV) and Incremental scale-up. The second barrier is Strategic objectives and its proposed mitigation strategies is: Using information to gain a competitive advantage. The proposed mitigation strategy for the third barrier, Degree of risk-taking, is Renting with an option to buy.

The applicability of the designed framework in a real-life setting is yet to be tested. Future research could involve using the framework and applying it in an actual implementation of IoT fullness sensors in a waste management organization.

With the advent of a smart society and an era of connectivity, there remain a numerous amount of challenges yet to be solved. One of the key IoT innovations analysed in this research is known as Unmanned Aerial Systems (UAS). UAS, or drone technologies, allow carrying out repetitive and dangerous tasks with almost no human intervention or supervision (Fernández-Caramés, Blanco-Novoa, Froiz-Míguez & Fraga-Lamas, 2019). The advent of drone utilisation in both public and civil domains has led to application areas such as realtimemonitoring and surveillance, parcel delivery, search and rescue missions (e.g. ER), remote sensing in agriculture and multiple other application domains (Valavanis & Vachtsevanos, 2015). However, the fast pacing development of interconnected devices and systems has outrun the human understanding and experience of usage, imposing challenges with regards to technical security, trust and of course privacy (Coetzee & Eksteen, 2011). Subsequently, most drones are limited to computing, storing and sharing data, making them vulnerable to adversarialattacks. A need for a more responsible adoption of drone-use exists and must be fulfilled with care. To this, a blockchain-based solution is proposed.The focus of this research is on a Dutch independent agency part of the Ministry of Agriculture, Nature and Food Quality, called the Netherlands Food and Consumer Product Safety Authority (NVWA). They have been actively using drones since 2017 and their main tasks consist of supervision, risk assessment and risk communication with the goal to protect the human and animal welfare (NVWA,n.d.). An explorative research approach was taken whereby the current bottlenecks experienced by the NVWA were identified and visualised, by means of desk research and semi-structured interviews. Subsequently, the sociotechnical effects of the use of drone detecting services by the NVWA were identified to obtain the key values for a responsible redesign. In this way, the role of the values, an understanding of the roles of institutions realizing these values, and the stakeholder engagement is understood. After identifying the open issues and conflicting values, a redesign phase began with the trade off between three mitigating security techniques.Blockchain technology is found to be the most suitable security technique in this research and fills in the ’lack of transparency’ gap which is crucial for a responsible redesign. Although it hasn’t reached full maturity and could induce undesired delays, it increases the security level of the organisation significantlyand takes into account the key values security and privacy. Moreover, the data managing method allows users to validate, maintain and synchronize the content of a transaction ledger which is replicated across the other users in the network (Tapscott & Tapscott, 2017).The solution has proven to work in existing literature and automotive contexts and is therefore generalisable in multiple on-going projects of the NVWA. Subsequently, the novel data managing method could be used as for revenue models such as digitised inspections. This indicates future growth options andpotential competitive advantages. All in all, the benefits outweigh the uncertainties of this technologyand it is believed that a blockchain based solution will be the next big step to achieve more responsible implementation of drones.

The sustainability of the Dutch healthcare system is under severe pressure with increasing expenditure, more demand for care and a shortage of healthcare professionals. In the current healthcare system, providers are rewarded for increasing volume, but not for adding value, which only exacerbates the situation. Value-based healthcare (VBHC) is a candidate framework to reform the current healthcare system into an outcome-based system. Information systems are essential for the implementation of VBHC to execute outcome and cost measurements and collaborate across the care cycle. A literature review shows that there is a knowledge gap in the scientific literature to the architectures and components for such value-enabling information systems. This thesis aims to fill that gap by identifying and designing the essential components for a value-enabling IT architecture for the Dutch healthcare system.
The Information Systems (IS) research framework of Hevner et al. (2004) is adopted to guide the research. The environmental analysis shows several deficiencies in the as-is architecture for VBHC including a lack of patient-centredness, inaccessible or unavailable data, and complex to extract and integrate data. Preliminary semi-structured interviews reveal that, besides IT barriers, there is a deeply rooted trust issue among stakeholders which hampers successful implementation of VBHC. It is found that trust is often related to transparency and can be managed. Increasing transparency in the healthcare architecture would allow for (i) continuous quality improvements, (ii) improved decision-making, (iii) positive financial stimulus and (iv) patient empowerment. The design activities aim to facilitate VBHC through integrating components that reduce or eliminate the IT barriers and increase transparency. Another round of semi-structured interviews is carried out to find the principles, requirements and components for the design. The overall design developed uses three types of environment: a Healthcare Information System (HIS) used by caregivers, a Personal Healthcare Environment (PHE) used by patients and a Quality Registration System, available to all stakeholders. The design and components are evaluated with an expert panel. The most essential components are: PHE, HIS, Quality Registry system, Clinical building blocks (Zibs), terminology standards, a data integration centre, logging and monitoring services, measurements and evaluations, auditing IT and care processes and an intermediary that stimulates improvement and collaboration. The expert panel reached a consensus that transparency in healthcare would contribute to solving some of the problems, but it will not solve the core problems that lie within the healthcare structure. There are several areas recommended for future research. First, the components should be evaluated with a broader expert panel to increase the validity. Second, the PHE is an essential component, but still in its infancy. It is recommended to further evaluate the utility for VBHC. Third, blockchain technology might enhance the design due to its inherent characteristics that offer transparency. Fourth, further research on the privacy-utility trade-off in the context of VBHC is recommended. Finally, interoperable systems rely on a shared and uniform language. It is recommended to research which standards have the highest potential to facilitate the architecture components.

Information technology innovation has given worldwide access to each other, to information, and to entirely new digital platforms forming new markets over the last decade. Multi-sided digital platforms are platforms that mediate between different groups of users, both buyers and sellers.
With IT the multi-sided platforms are exploiting the economic phenomenon known as network externalities, increase the users, increase the value of the platform.
However, the cultural differences between countries can make or break a successful entry into another national market. Simply copying the monolithic algorithms has not been the answer, as recently admitted by Amazon's departure from China.
This project looked at seemingly similar cultures (the Netherlands and Germany) to identify if the macro-level cultural dimensions influenced the acceptance and use of multi-sided platforms. This was examined through a survey that assessed the acceptance and use of online food delivery platforms, a prime example of multi-sided platforms.
The cultures do have significant differences, but their impact on the acceptance and use of online food delivery platforms was non-significant. The acceptance of online food delivery platforms still followed a 10-year different diffusion, but the significant differences on macro-level cultural dimensions did not relate to this acceptance.
An interesting conclusion to this project is that the countries, expecting to have similar cultures, are significantly different and the data gathered proves to be more substantial than this project could investigate. The UTAUT2-model did
show a significant acceptance and use of online food delivery platforms, which shows that the online food delivery platform has successfully expanded cross-culturally in the case of the Netherlands and Germany.

Agriculture is a fundamental element of every economy. However, global issues such as climate change, land deterioration, and a continuously growing population are strongly impacting the sector. It is estimated that by 2050 the population will increase by two billion, reaching 9 billion people to be fed. In the past 34 years, researchers investigated the use of potential technologies such as Artificial Intelligence (AI), Internet of Things (IoT), Machine Learning (ML), robotics, etc. The readiness of these technologies and the positive impact that they have on the sector is proven by numerous studies. However, these technologies are still not diffused enough within common farming practices. Adoption rates are very low, as well as full understanding of the technologies from the farmers’ perspective. Without a shift towards the digitalization of farming practices, the agriculture sector could be damaged, impacting both the future of society and general economy. This is why a change in the actual regime of production is needed: from traditional agriculture to AgTech. In this work, through interviews with winegrowers and analyses of technologies proposed by vineyards start-ups, an understanding of future market development is derived. The concept of sustainability is investigated as one of the major drivers towards the change of regime. More specifically, the study has been conducted on the viticulture domain, tackling in a very context-specific manner the problems concerning this branch of the field. A set of to-be-used criteria for start-ups business model creation is derived by the main needs of the growers and the degree of technology adoption observed. The criteria, or archetypes, have been developed on the base of the Sustainable Business Model Archetypes. A new model is proposed, which is a starting point towards the shift of regime and the digitalization of the agriculture field.

Recent developments in technology innovations show that huge efficiency improvements can be made in the manufacturing industry. Moreover, companies that adopt the new innovations, associated with Industry 4.0, can create a huge competitive advantage. However, these rather conservative businesses are slow adopters and usually wait for proof-of-concept before actual implementation. Since this industrial revolution, Industry 4.0, is still in its infancy, more research is required to get to full adoption. Although the industry still awaits proof-of-concepts, many different case studies have been performed, and with success! These clearly exhibit the versatility of the Industry 4.0-philosophy, makingwidespread adoption just a matter of time. One of the identified reasons for this lag in adoption is the lack of clear implementation guides despite the thorough research and redundancy of technology. Due to the holistic Industry 4.0-concept, many practitioners lose sight on how and what to implement. Various researches proposed the creation of a widely applicable implementation model, but this is yet to be developed. One of the prominent issues related to creation such model is the all-encompassing nature of Industry 4.0; it includes novel innovation in supply chains, in factories, and even in the products manufactured. Since there are clear differences between these ’applications’, a generic overarching model seems unreasonable considering the immense amount of variables to consider. This thesis depicts the first ever-made implementation model specifically aimed at improving raw material supply & planning in complex manufacturing companies.Supply & planning processes are the closest connections between a manufacturers’ own operations and its closest neighbours in the supply chain; i.e. suppliers and customers. Tapping into this specific field of operations enhances the utilization of Industry 4.0 both on the supply chain and manufacturing aspects. Through answering the main question, and several subquestions, relevant information is gained that enable the construction of an implementation model. The design of such implementation model includes a step-by-step approach for practitioners of manufacturing companies, and a clear description on what to consider at each step. Creation of the artefact (i.e. implementation model) happens by explaining the research question: How can Industry 4.0 be implemented into supply and planning departments of complex manufacturing companies using an implementation framework?By means of a design science research methodology (DSRM) the Industry 4.0 supply & planning implementation framework is designed. Through 6 pre-determined steps; (i) problem identification, (ii) objective definition, (iii) design & development, (iv) demonstration, (v) evaluation, and (vii) communication, it ensured that all relevant stages are included to construct a scientific substantiate artefact. Three of these elements in particular were considered to be main constructs of the thesis report. Through the objective definition stage, qualitative research in the form of interviews and literature review imposed what had to be included in the implementation model. In the design & development stage this information was casted into a mold, thereby being the first result to the thesis’ ultimate goal. The demonstration phase was assigned to check the applicability and effectiveness of the model by putting it into practice. Altogether a significant base of information was collected, obtaining the firstconceptual implementation model for Industry 4.0. In the existing tight markets in which various manufacturers operate, the utilization of improvement technologies is high. Techniques derived from methods like Lean, Agile and Six Sigma are used on a daily basis. Because companies are familiar with the use of these models, the adoption of newer versions becomes straightforward. Consequently, the implementation model is a derivative of such method, namely the DMAIC (Define, Measure, Analyze, Improve, and Control). Since these overarching steps do not provide sufficient information for actual implementation, extra delineation is applied through a combination of the Continuous Quality Improvement model and practitioners’ experiences. Via combination of the two, a first model consisting of 11 steps (i.e. within the 5 DMAIC stages) was constructed.The implementation model starts with goal identification, in which the companies’ digital transformation (i.e. Industry 4.0-adoption) strategies are adapted to local needs. Subsequently, the business processes are investigated thoroughly. By clever modification of an existing model called RAMI (ReferenceArchitecture Model Industrie 4.0), a standardized approach for identifying the key aspects of the business processes was obtained. Using the results of this business process modelling allows to diagnose the so-called key variables that have a considerable effect on the performance of operations. The top five of these key variables provide the focus for the execution of the consecutive steps. Data and information about these 5 variables is gathered through a process of replacing paper forms by digital forms and through connection of existing Operational Technology (OT) systems with Information Technology (IT) systems. Once all the relevant data for the five variables is obtained, a data analysis follows. Examining the inconsistencies in this data pinpoint the location where data enhancement (i.e. Industry 4.0-adoption) will significantly improve the process. Defining the performance indicators then help to know the business’ existing performance and allow comparison with future results, but also help users to monitor real-time process-efficiency by means of a dashboard. According to the Key Performance Indicators (KPI’s) chosen, technology introduction can finally happen. Thirteen different enabling technologies were identified during the literature research, providing practitioners a wide portfolioof options in their Industry 4.0-implementation. Shortly after implementation follows continuous monitoring according to the aforementioned KPI’s. By carefully assessing the business process’ performance, improvement studies can be performed and actual improvement of the system can take place. In the final stage it is evaluated whether the implementation was effective and what lessons-learned should be brought to the next technology-implementation. To test whether the implementation model indeed fulfil its vows, a test run is performed at an agriculture fertilizer manufacturing facility that definitely classifies as a complex factory according to the definition of this thesis (i.e. large portfolio of products and raw materials). The first few stages were quite obvious in their execution, mainly because of the clear instructions given. Especially the modified RAMI model gave useful insights and abandoned the requirement of complete Business Process Mapping which is very time consuming. Various key variables were obtained using a quality team. Since the majority of data -for these key variables- was already available, it was only a minor effort to obtain the rest using either OT-IT merger or digital reporting. In the case study, the data analysis stage was the most demanding task in both time and extra investigation. After describing the KPI’s related to the data analysis and describing the technology introduction stage, the real version of the case studyhad come to an end due to time and resource limitations. Continuous monitoring, improvement, and evaluation were further concluded through the sense of ’modelling’, where providing examples and describing the expected outcomes served as enclosure of the first trial. Although the model was designed with extra care and the input from both the literature review and the interviews were significant, some limitations still apply. It was observed that some of the stages were not definitive enough, making the actual goal of each step rather vague. As a result, some stages could take considerably more time than necessary, diminishing the model’s effectiveness. Moreover, the power of the data analysis, as described before, was truly reliant on my experience in statistical and data analytics. Therefore, the current data analysis-description requires more attention to advance the usefulness of this stage regardless of the users’ experience. Finally, the effectiveness and generalizability of the model were only touched upon briefly and require more in-depth investigation before claiming its novelty.

Predictive maintenance (PdM) is one of the promising technologies coming along with the fourth industrial revolution being pushed by disruptive technologies like Internet of Things (IoT), Artificial Intelligence (AI), robotics and Augmented and Virtual Reality (AR/VR). Adopting PdM potentially allows companies to reduce equipment downtime, increase the safety of their processes, increase revenue and develop additional business models. Although the promises of the technology are extensive, the successful adoption rate of this technology is still relatively slow. This is stemming from PdM’s multi-disciplinary nature and “hype” that over-promised its ease of implementation. Organizations are now starting to understand what is needed for efficient implementation, and this helps to manage the expectations about this technology. The fundamental problems highlighted in this research are the complexity, unclear vision, lack of knowledge and know-how in adopting AI predictive maintenance technologies inside an organization. According to Bain & Company’s survey companies in the industrial sector indicated that implementing IoT inside their organization proved to be more complicated than anticipated (Schallehn, Schorling, Bowen, & Straehle, 2019). There is a knowledge gap in the scientific literature, where a lack of best practice methods in terms of predictive maintenance implementation can be identified. Based on the problem highlighted and knowledge gap, the main research question was formulated: “How to facilitate the adoption of Artificial Intelligence-based predictive maintenance technology in the manufacturing industry?“. This study follows a phase-wise approach to obtain the research results. In the first phase, a literature study is conducted to identify the current situation about PdM, what information is available about the factors affecting this technology’s adoption and where is the knowledge gap to be filled. Selected factors to focus on with this research are discussed and agreed upon with the researcher and supervisors. In the second phase, the development of the best practices checklist is commenced. The centrepiece of this phase and the research project overall is the set of semi-structured interviews with 11 industry experts with extensive domain knowledge about predictive maintenance to collect best practices in PdM implementation. The insights gathered from the interviews are analysed in-detail in multiple iterations and then that filtered, aggregated information is used to develop the predictive maintenance project reference checklist. In the third phase, expert panel evaluates the practical applicability, generalizability and the validity of the constructed PdM checklist. Efficient implementation of PdM inside the organization could face numerous barriers and difficulties. Most of these barriers related to technologies using big data could be divided into three categories: technical, organizational and people related (S. Li, Peng, & Xing, 2019). Addressing all of these barriers in those 3 major categories would be unwise since that would not provide sufficient depth of analysis for each one of them. Selection of barriers is based on 3 criteria: the barriers must be relevant and applicable to the adoption of the PdM technologies; there should be a noticeable knowledge gap about how to overcome the barriers; the barriers must be complex enough (affecting multiple layers and stakeholders of the organizations) to fit with the Management of Technology multidisciplinary problem-solving perspective. Based on the information from scientific literature and consultancy reports on PdM, 3 relevant barriers to be focused on are chosen: business case building for PdM; trust in AI-based PdM (lack of trust in big data analytical results) and data management for PdM (the challenge of collecting the data, utilizing it and making sense of it). The interviews with the industry experts revealed valuable insights about predictive maintenance adoption, factors affecting the implementation and best practices that other companies have followed during the process of PdM realization. The most notable best practice that all the interviewees mentioned was involving all the relevant stakeholders early on. In addition, taking small steps, maintaining PdM platforms, celebrating small successes, showing a broad picture and providing a range for PdM business case were outlined. Furthermore, key factors that emerged from the conducted interviews influencing PdM adoption are delineated and summarized in this research project. These are useful for both practitioners and academic personnel who have an interest in this domain and want to gain further understanding of the dynamics surrounding predictive maintenance projects. This research project developed best practices reference checklist for predictive maintenance project implementation that supports organizations on high-level in adopting this novel technology by illustrating and bringing awareness to best practices that other organizations have been following during PdM implementation. This reference checklist is constructed to be a holistic, high-level PdM project support tool for the stakeholders proceeding with predictive maintenance implementation for the first time. This means that a detailed analysis of separate nuances is not sought after since that would misalign with the goal of being a wholesome, comprehendible overview of PdM project implementation checklist. Having a clear, structured and holistic perspective allows stakeholders to conveniently follow this checklist commencing and during predictive maintenance projects without being overwhelmed by excessively detailed information. This best practice checklist based on empirical study comprises a five-phase approach where the enablers and barriers in each phase are mentioned and suggestions on how to deal with them are outlined. These 5 phases are as follows: concept, feasibility, data, PdM algorithm development and operation phase. Furthermore, high-level, structured steps in each phase are laid out to support and offer recommendations to organizations with their PdM activities. In the end of each phase, an overview of best practices and barriers is delineated to recapitulate. In the concluding section of this best practices checklist, a compact, five-page adaptation of this reference checklist is devised for a quick overview of this constructed PdM project support medium and it is advisable to resort back to phases in the checklist itself if the more detailed explanation is needed. This compact version is meant for practitioners in the industry who have strict time limitations and wish to receive information quickly in a condensed format. To the best of our knowledge, such kind of high-level compact overview to assess PdM projects was not existing in the scientific literature. This research project directly investigates and provides a best practices checklist to fill this gap. In addition, this research provided design improvement ideas for different stakeholders to incorporate in their processes/products to facilitate better adoption of PdM. Trust factors affecting the implementation process of predictive maintenance are also outlined, helping companies to better communicate with their clients and internal organization about the benefits and usefulness of PdM. The developed research output has been preliminarily validated and evaluated by the expert panel that concluded that this best practice checklist indeed supports organizations in adopting predictive maintenance technologies. Furthermore, it was agreed that the output is clear and understandable with a well-structured approach. Coming from the high-level nature of this research, experts agreed that this research is generalizable to other industries. Main recommendations (for future research) include validating the best practice checklist in practice with multiple organizations inside the industry to correlate usage of this approach and success factor of implementing PdM. Furthermore, the development of additional support tools and frameworks to facilitate efficient implementation of predictive maintenance technologies would yield increased adoption rates of the technology. This research highlighted important factors contributing to the adoption of predictive maintenance technologies from organizational, people and technology perspectives. This helps to create more awareness about what is needed to consider for better adoption of this technology. Furthermore, a high-level structured overview of best practices checklist supporting PdM implementation is contributed to the scientific and practical domain, filling the previously outlined gap in the literature. In addition, coming from the analysed literature, this research complements the scientific literature on the topic of predictive maintenance by providing original content and additional awareness to the overall academic context regarding the dynamics of this technology’s adoption.

Edge computing extends the cloud computing capabilities to the edge of the network to facilitate processing of the data in the close proximity of its generation. It augments the deployment of several applications in the Internet of Things (IoT) domain which demand low latency and near real-time response for their reliable operation. However, the existing approaches that help accomplish this are inherently static and suffice only for scenarios offering a fixed service at the edge. Thus, rendering them infeasible for a large scale IoT scenario with several heterogeneous services and end users such as a Smart City. Among the several challenges in realizing edge computing solutions for such a scenario, the impact of incorporating key aspects of the interactions of the IoT devices and the clients with the edge, remains to be explored. These interactions translate to the ease of use which is vital to determining the degree of adoption of an edge computing infrastructure by the end users and subsequently its profitability for the service provider. This thesis presents EDIRO, which is an edge-driven distributed orchestration framework for edge computing that enables the edge to drive the workload orchestration through the collaboration of multiple constituent edge nodes. It takes into account the existence of on-demand and ephemeral nature of workloads that require an input, i.e., an IoT resource such as an image or sensor data for their execution. These IoT resources are sourced from the end users and the IoT devices in the vicinity. Recent studies that highlight the idea of collaborative processing and the simultaneous presence of data producers and consumers in an IoT ecosystem, support this vision. The underlying concept in EDIRO is the utilization of such IoT resources that are contributed by the end users in the vicinity, to carry out service orchestration for the client requests. To the best of author’s knowledge, this is the first work in the edge computing domain to conceptualize the idea of edge-driven distributed orchestration and implement a proof of concept followed by its evaluation and practical feasibility analysis. The main contribution of this thesis is the edge computing orchestration framework EDIRO, which is developed in Golang and released as open source to encourage collaboration with the community. Experiments are conducted on three different types of computing devices emulating edge nodes in a field scenario to determine the practical feasibility of EDIRO. The measurements include the time to serve a client request, the overhead due to the distributed orchestration approach and the computing resource utilization under bursty and normal traffic scenarios. The evaluation suggests that EDIRO is feasible for practical IoT use cases and provides a reasonable trade-off in terms of the benefits offered by this edge-driven approach and the overhead incurred. This thesis shares valuable insights into the ways in which this work opens up the scope of further research in this domain along with the key findings from the system development and experimentation phase.

Autonomous vehicles have advanced rapidly in the last decades. The automotive industry is making valiant efforts to strengthen autonomous vehicle technologies and improve overall road safety. Besides actual safety, feeling safe is equally important. In other words, there is also a socio-technical perspective of self-driving vehicles that should not be ignored. Currently, when pedestrians wish to cross the road, they tend to rely on informal communication with drivers. However, this will not be the case with fully autonomous vehicles as the driver or passenger will most probably execute various tasks and will not pay attention to the road. Consequently, this master thesis investigates the loosely coupled relationship between pedestrians and fully autonomous vehicles in crosswalk use cases with no traffic lights. The main research question that is investigated is "How should the interaction between humans and fully autonomous vehicles be established in order to influence road safety and road awareness?". To address this research question, this master thesis espouses notions used in the scientific field of Interaction Design and applies the goal-directed design approach.

The Internet of Things (IoT) will soon impact the lives of thousands of people as numerous IoT devices are emerging in the consumer market. Consumers goods consist of products designed for the consumption of final consumers. Even though IoT applications are expected to improve people's lives, security is often lacking in current IoT devices. Vulnerabilities in these type of products pose serious risks to the security and privacy of consumers. Compared to traditional electronics, IoT devices are endowed with internet connectivity that can be exploited by hackers in remote attacks. Several attacks on IoT products that can threaten the security of a large of number actors have already been observed. To minimize the risk of attacks, developers and vendors need to identify vulnerabilities in time before any malevolent individual can exploit them.
In recent years, as part of vulnerability management practices, many organizations have started to implement crowdsourced security methods such as Bug Bounty Programs (BBPs) and Responsible Disclosure Policies (RDPs). BBPs and RDPs are programs that involve the participation of ethical hackers in the security processes of organizations, reporting vulnerabilities to companies in exchange for monetary rewards or recognition. These methods present the benefit that thousands of hackers can work together with companies to identify and patch vulnerabilities. Empirical research suggests that BBPs and RDPs effectively augment existing vulnerability management practices by companies. However, the application of these programs in the field of IoT has never been studied. There are many questions open regarding the potential and future adoption of Bug Bounty Programs and Responsible Disclosure Policies. The research aim is to study and expand the literature on security practices for IoT, focusing on the application of BBPs and RDPs, and to conduct an interview-based investigation with experts in order to provide practical recommendations for companies to enhance vulnerability management practices for IoT consumer goods. For this research, the literature on IoT security and security practices is confronted with empirical data from expert interviews. The empirical data was gathered during an internship at Deloitte in the Netherlands. In total, 19 interviews with cybersecurity experts from different companies in the field were collected for this thesis. The results are employed to generate recommendations for companies to improve their vulnerability management practices with the use of BBPs and RDPs. The recommendations are directed to companies developing, manufacturing, and commercializing consumer IoT devices that want to enhance the security of their products. The main contributions of this research consist of practical and tangible security recommendations for companies to tackle IoT vulnerabilities in consumer goods, which will help enhance the overall IoT security practices. Moreover, our findings raise attention on the societal risks derived from the unsafe deployment of vulnerable IoT products into the consumer market. We create awareness on the IoT security challenge, and present a call for further actions from companies, consumers, and regulators in the IoT domain.

Edge computing can deliver substantial value to the general idea of the Internet of Things (IoT). However, there is a myriad of potential IoT applications for edge computing. Stakeholders are left with uncertainty about how the business potential of edge computing for these IoT applications can be identified. This research contributes in solving this, by designing a business model tool that can be used to identify the business model potential of edge computing for distinct IoT application areas, based on business model viability and feasibility. Through the Design Science Research Methodology (DSRM), the tool has been designed, demonstrated, and evaluated. Based on the STOF ontology, and supplemented by the theoretical domains of business ecosystems and platform theory, nine generic variables have been identified to explain business model viability and feasibility. These generic variables have in turn been contextualized towards the edge computing domain, in terms of 45 contextual input variables. This is the first research that unfolds these business model variables for edge computing.