The Topic

The constraint-based product configurator CAS Merlin has a large and ever increasing set of parameters that have a huge impact on its performance. Setting the proper values for these parameters is very complex and time consuming and yet highly important, especially for large workspaces with complex product structure and many rules. Currently, the parameter tuning is done manually by trial and error. Fortunately, there exist algorithms that can automate this process. The goal of the thesis is to identify such an algorithm for the product configurator Merlin and to implement it.

Your Tasks

• Conduct a literature research to find a fitting automatic parameter tuning algorithm for Merlin.
• Customize and implement the selected algorithm in Merlin.
• Experimentally evaluate the automatically tuned Merlin against Merlin with current/default parameter settings on real-world customer data.
• Summarize the results in your thesis and/or scientific publication(s).

Your Skills

• Ability to read and understand scientific papers.
• Next level JAVA coding skillz (so that the programs you write run fast).
• Communicating your ideas clearly and precisely in English.
• Working together with others in German is also easy for you.