The Data Science chair carries out practice-based research into the creation of data products. It focuses particularly on products that are important in a delta area such as Zeeland, for example for coastal protection, safety, tourism, food, industry, energy and logistics. The lectorate works together with entrepreneurs, governments and other lectorates and knowledge centres so that data products can be combined with knowledge from these domains.
With data products, users can take faster, more efficient, more effective and more accurate decisions. Disciplines such as mathematics, statistics, software engineering and machine learning come together in Data Science.
Business understanding
Every data product is based on a research process. In the first, crucial step (business understanding), the question is investigated.
After that, raw data is collected and made suitable for the rest of the process. A possible next step is to visualise the data or communicate it in another way. The data can then be further modelled using machine learning. This includes various techniques such as clustering, predicting, classifying and detecting anomalies. Ultimately, this leads to a data product.
Mischa Beckers is a lecturer in Data Science. He delivered his lectoral address on 16 November 2017: 'Do believe the hype'.
Innovative solutions to practical issues, such as preventing repairs and sustainable reuse of equipment, are being sought by carrying out practical research and demonstrations. It looks at reusing electric motors and transformers when renewing installations, sharing little-used maintenance tools and new cleaning techniques that drastically reduce waste water and the use of chemicals. Furthermore, new, circular business models are being developed for the maintenance sector of the process industry. In each case, the key question is whether a solution is also economically viable and, at the same time, practically feasible for companies. The project was funded within the Interreg V programme Flanders-Netherlands, the cross-border cooperation programme with financial support from the European Regional Development Fund.
Not only in Zeeland's municipalities, but also outside Zeeland, asset management tools and data science are underused in the management of sewers and pumping stations. Combining asset management and data science, Riobase aims to provide validated action perspectives for operators to make the management of their sewer systems more effective and efficient. Mutual cooperation for knowledge development, information sharing, planning and experience are central to this. Appropriate instrumentsIn the project, a draft framework with appropriate tools for asset management was developed with the managers. This is being tested in the municipalities of Reimerswaal and Veere. Then there is a second round in two other municipalities. The result of these tests is a validated framework with tools that works in a small municipality and enables managers to carry out risk-driven and data-driven sewer management. The framework also includes a (geo)visualisation with static and dynamic data to support mutual decision-making. Education and practiceThe research feeds into teaching at HZ's Civil Engineering and Water Management programmes, Data Science Track of HBO-ICT and Technical Business Administration. An example of translation into professional practice is a practical training course for administrators. The results can also be used by administrators in other municipalities of similar scale.
