Refine
Document type
Language
- English (11)
Keywords
- Internet of things (11) (remove)
Course of studies
Ensuring data quality is central to the digital transformation in industry. Business processes such as predictive maintenance or condition monitoring can be implemented or improved based on the available data. In order to guarantee high data quality, a single data validation system are usually used to validate the production data for further use. However, using a single system allows an attacker only to perform one successful attack to corrupt the whole system. We present a new approach in which a data validation system using multiple different validators minimizes the probability of success for the attacker. The validators are arranged in clusters based on their properties. For a validation process, a challenge is given that specifies which validators should perform the current validation. Validation results from other validators are dropped. This ensures that even for more than half of the validators being corrupted anomalies can be detected during the validation process.
The importance of machine learning (ML) has been increasing dramatically for years. From assistance systems to production optimisation to healthcare support, almost every area of daily life and industry is coming into contact with machine learning. Besides all the benefits ML brings, the lack of transparency and difficulty in creating traceability pose major risks. While solutions exist to make the training of machine learning models more transparent, traceability is still a major challenge. Ensuring the identity of a model is another challenge, as unnoticed modification of a model is also a danger when using ML. This paper proposes to create an ML Birth Certificate and ML Family Tree secured by blockchain technology. Important information about training and changes to the model through retraining can be stored in a blockchain and accessed by any user to create more security and traceability about an ML model.
Dealing with Data Quality in Smart Home Environments — Lessons Learned from a Smart Grid Pilot
(2016)
The buzzword “smart city”, which describes the integration of digital technologies in different areas of cities, is on the rise worldwide. Even though there are various megatrends pushing towards the uptake of smart cities, both public sector and businesses struggle to indentify viable sources of financing and business models for smart city initiatives.
This also holds true for the smart city business of Bosch Software Innovations GmbH, which includes various smart city projects in initial stages. Therefore, the topic of business model development in the context of smart cities is being approached in this thesis, with the Gambit project in the City of York being examined as a reference project. The idea of Gambit – “Gamification for better living in cities by influencing tourist behaviour” - is based on a smartphone application for tourists, which aims to influence visitor behaviour through elements of gamification. Thereby city services should be assisted in mitigating the problem of local overcrowding in the city centre. In its initial phase, the project is financially supported by public authorities. The central question is how such a kind of project can be financially viable and how its commercial uptake can be achieved. With this in mind, the aim of this thesis is to explore business model development for smart city solutions based on the example of the Gambit project in York. To do so, a multi-method approach is used, comprising a literature review on the theoretical background of smart cities and business models, as well as empirical research based on interviews with the partner organisations of the project, as well as a workshop with other city stakeholders.
This key findings show that the attraction of additional stakeholder within cities is essential for the economic success of smart city projects. This implies multi-directional value streams and multiple sources of financing within smart city initiatives. In this context, various forms of financial contribution, such as indirect payments through other offerings and the provision of advertising space to co-finance solutions should be considered. Besides, linking smart city solutions to other services within cities might form the basis for financial viability. The findings produced stress the importance of collaboration and partnering. Furthermore, it can be concluded that the diversity of stakeholders implies a diversity of value streams in smart city business models.
Testing applications for SmartHome environments is quite complicated, since a real environment is not accessible or the conditions are not controllable during development time. The need to set up the whole hardware environment, increase the complexity of these systems enormously. Therefore, it is helpful to simulate the SmartHome hardware components and environment conditions (e.g. rain, heat, etc.). This paper contains an approach to improve the test and demonstration process of Internet of Things scenarios. A prototype (ScnSim: Scenario Simulator) was developed to set up scenarios. The user of the ScnSim can create her/his own scenario using items (sensors/actuators) and rules, which control the sensors and actors building the IoT enviornment. This simulator is supposed to support the user testing IoT applications or configurations of SmartHome platforms like openHAB. In addition, the ScnSim is supposed to help demonstrating showcases, for example, often demonstrated on a trade fair or as a proof of concept for a customer.