HZ Stern Thesis prize 2020: In-depth search is rewarded!

Kees Bal, editor HZ Discovery

The HZ Stern was awarded from the HZ studio in Vlissingen on Wednesday 9 December 2020. This year, two students may call themselves winners of this prestigious prize: Miriam Günter and Omar El Nahhas. HZ Discovery investigates their excellent research.


The nominees present themselves and their research here in a video:

  • Roelant de Looff, ICT: “Optimizing through big data”
  • Maximilian Foy, Civil Engineering: “Modelling the loading process for subsea cables”
  • Kim van der Vliet, Business Administration: “GGD Zeeland and the mismatch between information provision and information need”
  • Miriam Günter, Vitality and Tourism Management: “Niche market exploration for enhancing quality tourism”
  • Omar El Nahhas, Engineering: “Classification of visually unrecognizable steel defects”
  • Maartje de Smit, Primary School Teacher Training: “Telling regional stories in the classroom”

Jessie Goossens, Social Work: “Motivating LVB clients in forensic psychiatric care” no video available

Miriam Günter: From ‘Kleinarbeit’ to a big result Miriam, a German student in Vitality and Tourism Management, has come to call Zeeland her second home. So being able to help a place she knows and really likes, was a great facilitator and motivator for her graduation work. She is thrilled with the Stern prize as it is a recognition for quality and for relevant, sense-making and practical outcomes after a year of hard work (see box 3 for Miriam’s research). Also the job market in the tourism industry will be limited in the next months, maybe years, so this prize may boost her profile.

As a rather detailed person, Miriam came up with a very thorough research approach, in which several models have been evaluated and connected. This process was a bit of a puzzle at moments (see the photo to the right), but it provided a solid basis on which to base the interviews and analysis. Strategic thinking and a love for tables, arrows, colour separations and excel did the rest. This is also what she is most proud of, the many things she did that were not really expected but added quality to the methodology.

Luckily she had all her interviews done when the tourism lockdown came. Analysis was tackled from the home office and emerging hypotheses were confirmed via email. The most important target market factor for low seasons turned out to be the geographic factor, the place of living. It left her thinking: “This is so simple… kind of exciting and unexciting at the same time.”

Miriam is keen to give fellow students a few tips for a good graduation. First, pick a topic you are passionate about and that can benefit more than just one company, Second, put a lot of effort into your research proposal! Then still you will always experience some difficulties during execution. Therefore, also have a plan B! Third, talk with all sorts of people, as it gives you new ideas and insights. And lastly, be creative in what you can do with your results. Whenever a smart idea comes into your mind: Write it down!

Miriam’s room: solving the puzzle

This is what 'Kleinarbeit' looks like

Miriam’s research: Niche market exploration for enhancing quality tourism What are the benefits that German tourists see in spending holidays in Zeeland during low season that make them travel there? The reason for asking this is that few tourists visit Zeeland in winter (while many do so in summer), which is difficult for local businesses that profit from tourists and reduces the attractiveness of Zeeland to live and work there all year-round. Thus, VVV Zeeland wanted to find markets of quality tourists that are unknown yet and could be targeted more by them for winter. Thirty six interviews were conducted with German tourists visiting Zeeland between 11/2019 and 2/2020 and 8 customer service employees in tourism, asking what makes Zeeland an attractive destination during low season. The interview questions were developed, structured and analysed combining two models.

Model 1 comprises of 5 decision steps that tourists walk through when deciding, what they do where during their holidays (Mathieson and Wall, 1982). Topics were current travel behaviour, tangible and underlying intangible benefits in low season, travel satisfaction and ideas of improvement, and personal questions (place of living in Germany, age, etc.)

Model 2 focuses on specific factors that influence a holiday decision (Horner and Swarbrooke, 1996). The most relevant ones for the VVV were identified and searched for in the answers of the interviewees. Furthermore photos of holidays in Zeeland were shown for triggering memories and emotions (photo elicitation), thus lowering the chance that they forget something in their answers.

Finally, all the statements of the respondents were grouped by tangible and intangible benefits (that was really “Kleinarbeit” - incredibly meticulous work - as they say in German), to identify clusters that could stand for potential target markets. Subsequently it was investigated whether those people who searched for the same or similar benefits had certain characteristics (like age or place of residence) in common.

Very briefly summarised, the conclusion was that the current market of North Rhine-Westphalia should be extended by the federal states of Hesse and Rhineland-Palatinate, targeting the following sub-segments:

  • Young families with children under school age;
  • Health tourists;
  • Winter campers;
  • ‘Explorers by chance’ who show interest into maritime topics and war history;
  • And families from federal states with winter holidays in February.


  • Mathieson, A., & Wall, G. (1982). Tourism: Economic, physical and social impacts. London, England: Longman.
  • Horner, S. and Swarbrooke, J. (1996). Marketing Tourism, Hospitality, and Leisure in Europe. London, England: International Thomson Business Press

Omar el Nahhas: Promise little, yield a lot It was not easy for mechatronics student Omar to find a graduation assignment focused on data science and machine learning. He eventually found a warm welcome at Tata Steel. Stimulated by the hype, he had already studied these subjects outside his studies through other courses. It turned out not to be magic, but mainly linear algebra and calculus (and hard work). The advice of the managers at Tata was to make little promises in his research proposal and then yield a lot, to show the contrary. And they succeeded. Physical proximity inside the factory was more difficult due to the corona measures but contact with higher management was easier. This led to a much better understanding of the bigger picture, the business case and the potential impact. That was a great motivation for Omar and also ensured the end product was actually usable in practice. Writing the thesis certainly had its ups and downs, but the result has shown to be very complete.

Omar was surrounded by people with a PhD in a science field. They taught him a lot in the analytical field and kept him sharp in the way of documentation. There were also consultants who helped him to create better presentations. For example, now he knows how to clarify the purpose of a slide at a glance. Since September Omar has been a master student in Digital Manufacturing, partly in Spain and partly in Latvia. His future plan is to start a business and find innovative, sustainable solutions to problems related to the seventeen Sustainable Development Goals of the UN. He is therefore always open to expand his network (via LinkedIn) and to conduct exploratory discussions, for example about the digital transformation of factories and Smart Industry1

1 Smart Industry or Industry 4.0 is the name for a trend of automation and data exchange that is used in industrial manufacturing techniques. It consists of autonomous cyber-physical production systems, the Internet of things, cloud computing and systems that can partly or completely take over our thinking process (cognitive computing).

Omar’s research: Classification of visually unrecognizable defects on steel coils Steel defects at the end of the process are visually unrecognizable, making it very difficult to determine the defect’s cause. But if this so-called root cause is not addressed, all subsequent roles will also have similar defects, all of which must be fixed. That takes a lot of time and money. Finding this root cause is a time-consuming job that was previously only done by an expert. A month of data therefore took a month of full-time work!

The research assignment was therefore to design a system that automatically classifies steel defects. For this, ‘computer vision’, a branch of machine learning, has been used that allows you to classify images/objects based on certain algorithms. By means of the V-Model, the entire conceptual system is broken up into subsystems with components. These components were then developed and tested, and reassembled step by step into a fully functional system. Close contact with project stakeholders was maintained throughout the project to ensure that every assumption and choice in developing the system was correctly applied in practice.

What previously took a month to analyze a month’s worth of data, is done by the new system within 45 minutes. That is an optimization of almost a thousand times. Now that the analysis is automated and it is economically feasible to apply it to all incoming data, Tata Steel can potentially save millions of euros per year by quickly analyzing the root cause. This system also enables preventive maintenance on a large scale.