Jordan Cork, Andrejaana Andova, Tea Tušar, Pavel Kromer and Bogdan Filipič

Abstract
Constrained multiobjective optimisation problems (CMOPs) are
common in real-world optimisation. They often involve expensive
solution evaluations and, therefore, it is helpful to know
the best methods to solve them prior to actually solving them.
These problems also tend to be relatively difficult for algorithms
compared to the majority of test problems. This difficulty often
presents itself in the infeasible region, calling for a focus on the
constraint handling technique (CHT). The purpose of this work is
to select the best CHT for problems with difficult constraint functions.
This first involves the collection of a set of such problems.
CHT selection is then conducted using problem characterisation
and machine learning. The outcomes are positive in that prediction
achieved a high accuracy. Additionally, further insights are
provided into the features that describe CMOPs.