It is an energetic Henrik Smedberg who shows up for this interview, and after a quick visit to the coffee machine we start our conversation. It was in mid-June that PhD. student Henrik held his half-time seminar. An element that all doctoral students go through, it focuses on progress, feedback and learning rather than results, assessment and control. The title of his research is "Interactive Knowledge Discovery for Knowledge-Driven Multi-Objective Optimization and Decision Support ".
Game development as a starting point
Henrik’s academic journey started when he was accepted to the bachelor degree program Game development and programming at the University of Skövde.
- I’ve always been interested in programming and algorithms. And after my bachelor I got a job as a research assistant working with optimization problems here at the University, says Henrik and continues. That is how I came in contact with many of the colleagues I have today.
Henrik’s work as a research assistant inspired him to continue studying and not long after he took his master degree in automation engineering which ultimately lead he where is today. The aim of Henrik’s research is to help decision makers make better decisions by identifying patterns that lead to the best performance.
Typically, decision makers have certain preferences that are used for selecting a final solution to be implemented in practice. Most multi-criteria decision analysis methods focus on the solutions’ performance in the objective space. However, it is in the decision space where practically relevant knowledge resides.
- Having access to this knowledge can help decision makers gain additional insights into the problem and the optimization process, leading to more informed decision making. The method I use can be applied to any optimization problem, says Henrik.
The car exemplifies Henrik’s method
- I can use the performance of a car to exemplify my research. We have horse power on one end, and a fuel-efficient car on the other. To understand the optimal performance, according to preferences, we need to know the relations between horse power and fuel efficient and the decision variables. This data is analyzed in a decision support system that I have developed, Mimer. A software that identifies these patterns that describe the relationships between the variables in the decision space and the outcome in the objective space, according to defined preferences. The result helps understand the best performance so the car can be optimized accordingly, says Henrik.
Enjoy doing research
Henrik is now half way through his PhD. and when asking about his plans for the future, Henrik is quick to answer.
-I’d like to keep working with research, whether it is within academia or industry . I really like doing research, it is fun, concludes Henrik Smedberg.
The decision support system Henrik uses in his research is called Mimer. It is developed by Henrik and allow him to understand relations between different performances.